走进Moderna自COVID以来最大规模的mRNA试验

For the first time, a cancer vaccine that works has appeared. But the industry has been working on this for over 20 years, more than 1,000 clinical trials have been conducted, and they have all failed. What was different this time? What is it about mRNA technology that allows the immune system to learn in ways that other approaches could not?

There are always cancer cells in our bodies. Our immune system is very well trained to spot these cancer cells in their early stages and get rid of them. But if the cancer grows, the question arises: How do we retrain the immune system?

We’re essentially taking a biopsy of your tumor. We’ll read all the letters of your DNA and then do the same with a healthy cell in your body. We’re literally going to compare them letter by letter, nucleotide by nucleotide. Then we’ll use an algorithm to determine which of these mutations are the most significant. When this is introduced into your body, it teaches the immune system to recognize the signs of a cancer cell that it missed.

How is this regulated?

Since each dose is individual, of course, each dose cannot be approved separately.

1. How Moderna Went From COVID to Cancer

I’m Jorge Conde, a general partner on the a16z Biotechnology and Healthcare team. Today, I’m very pleased to welcome back Moderna CEO Stéphane Bancel.

Those who have been listening to us for a long time may remember that Stéphane joined our podcast in December 2020. Then we talked about all the work Moderna did to give us the mRNA vaccine for COVID-19. At the time, if you go back and listen to that episode, you’ll hear how quickly Moderna was able to respond to the emergence of the virus, analyze it, and essentially “print” a vaccine that would protect people from COVID-19, thereby significantly improving our situation with the pandemic.

We called that episode “The Machine That Created the Vaccine.” The reason we’re talking again today, in August 2026, is that Moderna has announced a huge breakthrough in using mRNA technology to fight cancer.

So I want to pass the floor to you, Stéphane. Once again, we’re very happy to welcome you. Thank you for coming back. Perhaps let’s start with the most important thing—the news.

2. The Phase 3 News: mRNA Cancer Vaccine + Keytruda in Melanoma

Moderna and Merck announced in early August that they were conducting phase 3 clinical trials for the treatment of melanoma, which yielded promising results. So let me give you the floor to tell us what exactly you announced. What did you see during this third phase of testing? Then I want to delve into what Moderna is doing in the fight against cancer.

Jorge, thank you very much for inviting us again. We’re very happy to be here with you.

Indeed, last week, we and our colleagues at Merck announced that, after 10 years of working on personalized cancer treatments using our mRNA technology, the phase 3 trial was successful. This is the first result of many. This is the first time a melanoma drug has been shown to be more effective than Keytruda therapy alone, so this is extremely important for melanoma patients.

This is the first time a cancer vaccine has been proven effective. As you know, the industry has been working in this field for over 20 years. I think there have been more than 1,000 clinical trials, and if you look at the phase 2 data, we announced last week that we met the primary endpoint of the trial, which is recurrence-free survival. That means people do not face a return of the disease or death.

We achieved this, to our own surprise, as this was only the first interim analysis. This is not the end of the research; this is only the first interim analysis of the results. What came as a surprise even to us was that we achieved the secondary endpoint—distant metastasis-free survival—which takes more time to assess, as it means the absence of metastases from the primary tumor.

We achieved that goal as well, which was great. Again, this was unexpected for us, but it means that the results are really very good. We’ll soon be presenting the data at a major medical oncology conference, as is customary in our field.

To give you an idea of the magnitude, at the ASCO conference in the spring of 2026, a few months ago, we showed that the results of our phase 2 trial, also comparing it to Keytruda, showed about a 50% recurrence-free survival compared with those who received Keytruda alone. And this is 5 years after treatment.

As you know, in oncology, doctors consider 5 years to be equivalent to being cured. So this is an extremely big event. If you look again at the data from that phase 2 trial, you’ll see that about 80% of people remained healthy 5 years after treatment and surgery to remove the melanoma.

We’re really excited about what this means for the industry. We’re already working hard with regulators to submit the documents and make the drug available to patients as soon as possible, hopefully in 2027. We’ll strive to do this as soon as possible.

The Massachusetts factory is ready, and we’ve figured out how to manufacture and scale the product for each person individually in a timely manner.

3. What Is Keytruda & Why Isn't It Enough Alone?

There’s so much to sort through. This is a tremendous advance in treatment—in this case, melanoma—and hopefully, over time, it will become more widespread.

First, tell us, for people who are less familiar with cancer treatments, what Keytruda is and why Keytruda alone is not enough. Why was Moderna needed? Why was this particular mRNA cancer vaccine needed in combination with Keytruda?

Of course. Keytruda is one of the leading immunotherapies that people might have heard of. To simplify it for non-biologists, it’s essentially a molecule that opens the gate to let out the “dogs” that will attack your cancer using the immune system, if you will.

Checkpoint inhibitors, when they work, are fantastic because these people, 5 years after treatment, are cured in the sense that they have no disease left. But only 60% of people remain healthy after 5 years, according to the published data from the phase 3 study of Keytruda.

For those 60% of people, it’s amazing. But that means there are 40% of people who are undergoing treatment, fighting cancer, and the treatment doesn’t really help them. What’s also difficult is that these treatments are wonderful, but they often have very serious side effects.

Unlike chemotherapy or radiation therapy, the side effects of immunotherapy are mainly immune-related. If you look at the label for our clinical trials, people who get checkpoint inhibitors—whether it’s Keytruda or another checkpoint inhibitor from a different company—go on to develop type 1 diabetes, lupus, Crohn’s disease, and things like that.

Of course, it’s better to have such diseases than to die from cancer, which is why these drugs have become the standard of treatment. But think about the 40% of people who don’t respond to checkpoint inhibitors: They mostly get an autoimmune disease and don’t benefit from the drugs.

What we’re trying to do with Merck is develop a technology that originated in Moderna’s labs back in 2015 or 2016, when we partnered with Merck. We were looking for the best company in immunotherapy to collaborate with and have a complementary approach.

The idea we had at the time, using our experience with infectious-disease vaccines, was that we had learned a lot about the immune system and how mRNA interacts with it. We thought we could develop a mechanism of action that would be completely orthogonal—that is, completely different from immunotherapy—because you could start with the sequence of your tumor.

Then we can develop a product that essentially teaches—to go back to the dog analogy—teaches these dogs what to look for, very specifically. That’s the real beauty of our technology.

If you think about Keytruda, you’re essentially letting the dogs off the leash, but they sometimes act a little erratically. This is your immune system. Our product, by contrast, is able at the molecular level, inside your immune system, to teach your T cells very specifically: This is what you need to look for, and it is actually on your cancer cells. So these T cells go and essentially attack your cancer cells.

4. Preventing the Return of Cancer: What "Vaccine" Really Means Here

This is where personalized cancer vaccine technology will have a big impact on melanoma treatment. We’ll come back to the personalized part later, because I find it fascinating—not only from a technological point of view, but also from an operational point of view. I want to come back to that.

But let’s focus on the word “vaccine.” Usually, when you think of the word “vaccine,” it’s something to prevent disease. In this context, these patients already have cancer. In a sense, you’re preventing something: You’re not preventing the cancer itself; you’re preventing it from coming back.

That in itself is extraordinary when viewed from the perspective of therapeutic intervention. So, first of all, is this a fair characterization?

A fair characterization. The reason the word “vaccine” is used in this industry is that we didn’t coin it. We simply followed common practice.

As I mentioned, the reason there have been more than 1,000 clinical trials is that this approach is about training your immune system. If you think about getting vaccinated against COVID-19 or the flu, you’re training your immune system before the virus infects your body. Here, you’re teaching your immune system not about the virus, but essentially about the cancer signal that it missed.

What we know today in this field is that cancer cells are in our bodies all the time. External factors, or simply errors in cell replication, create DNA mutations that become cancer cells. Our immune system is very well trained to spot these cancer cells in their early stages and get rid of them.

But if your cancer is growing, the question arises: How can you retrain your immune system? I think that’s why the industry also uses the term “vaccine” for this approach.

5. Why 20 Years and 1,000 Trials Failed Before mRNA Worked

Even if, as you said, it is a therapeutic approach after the onset of cancer, it’s about training the immune system. As you rightly point out, the industry has tried to do this many times before. About 1,000 clinical trials have tested this theory. They all failed. You, Moderna, and Merck have succeeded here. What was different this time? What is it about mRNA technology that allows the immune system to learn in ways that other approaches have not been able to?

Yes, I think there are 2 components. One is mRNA technology, and the other, I believe, is individualization by creating a product for each individual. So let me talk about these 2 points.

With respect to mRNA technology, we know—and have published data, including with the Karolinska Institute in Sweden, which awarded the Nobel Prize in Medicine in 2015—that with our technology—I can’t speak for other mRNA companies that use other mRNAs, lipids, and so on—when we inject our mRNA into a muscle, whether it’s a COVID vaccine, a flu shot, or a cancer treatment, the mRNA goes into the lymph node and enters the APCs, the antigen-presenting cells, which, as you know, are a key component of your immune system.

The mRNA gets inside the APC. We demonstrated and proved this at one point together with the Karolinska Institute. Then you actually translate the message that’s in the mRNA inside the APC, inside your immune cells, and present it from the inside.

So we think this is a very important difference from most previous cancer vaccines, which were made from proteins or peptides in reactors and then injected into the patient. They actually get into the bloodstream because, as you know, when you inject recombinant proteins, they just get carried around in the bloodstream. Your immune system sees them, but not in the same way as mRNA, where the process happens from the inside. So we think this is a very important component of antigen presentation, if that sounds clear.

The other component is true individualization. Previously, many people tried to use technologies based not on mRNA, but on proteins or peptides, as well as using shared antigens. But here we assume that cancer is a disease of DNA. Because the cost of sequencing has dropped significantly over the past 20 years, we will take a biopsy of your tumor. We will read every letter of your DNA—all 3 gigabytes of her genes—and then do the same with a healthy cell in your body.

We’re literally going to compare them letter by letter, nucleotide by nucleotide. We will then use an algorithm to determine which of your hundreds of thousands of mutations are most relevant, drawing on current knowledge in immunology and oncology. We select 34 mutations that we consider to be the most important and combine them into 1 big mRNA molecule, which we manufacture for you in 30 days. It is then administered intramuscularly in the hospital.

When it gets into your body, it actually teaches your immune system to recognize the “signature” of your cancer cell that it missed. This is not a signature shared by other patients, but a very specific signature of your cancer cell. We showed this at ASCO and published the results of the phase 2 study, and we believe it will work in phase 3 because it is mechanistically sound.

Approximately 90% of antigens differ between patients. When we started, we had no idea what it would be like because the entire industry was based on ready-made antigens. At the beginning, we thought, “We don’t know if we’re going to get 2%, 5%, or 90% antigen matches in all patients.” In fact, 90% of antigens in humans are different. So the only way this can work is through personalization.

6. Inside the Algorithm: How Moderna Picks the 34 Neoantigens

Which individualizes it, exactly. It’s being carried forward. In that regard, if you compare a normal genome to a tumor genome, you find differences. I’m curious how you arrived at 34 as the correct number. I’m sure there’s a very good reason for this, but what algorithm allows you to do this? Is this Moderna’s own development, or is it something that is already known in the industry? Help us understand; help us look into this “black box.”

Of course. I’ll open this “black box” a little bit, but not too much, because there’s a lot of know-how and things that are very confidential for us.

Essentially, we started with what is already known in the industry. We use many databases and publications, as well as the expertise of many of the best scientists and doctors in immunology and oncology. With that starting point, we use a lot of internal data that we’ve collected over time.

We also collaborated with companies that, because they work in diagnostics, cell therapy, or other areas of oncology, had access to large amounts of data, T-cell mapping, and so on. That’s why they were useful for learning.

What’s interesting about the data we released last week is what I think of as “intismeran autogene 1.0.” The algorithm for the phase 3 trial was the same as in phase 2 and phase 1. Since we’ve been doing this for 10 years, this is a 10-year-old algorithm.

If you look at the data, it’s doing pretty well. As we talked about with the phase 2 data, 80% of people remain free of signs of the disease after 5 years. This is amazing for these patients. But there are still 20% of patients who do not respond to treatment.

One of the things that we’re going to do now that we have access to the data and samples from the phase 3 patients is go back and analyze that data to figure out why some patients responded and others didn’t. We have access to all their blood samples, their sequencing data, everything.

We’ll see if we can improve the algorithm, and we’ll go to the FDA if we find a scientific basis to change the algorithm—to go from, say, 1.0 to 2.0—and then we’ll change it. Of course, we have to do this in a very controlled manner so as not to lose effectiveness or, obviously, do harm.

But I think about it this way: When we talk about AI, we always joke that the current version of AI is the worst version we’ll ever see in our lifetime. The same goes for intismeran autogene. Moderna’s current version of intismeran autogene is the worst version you’ll see in the history of medicine.

This gives me a lot of hope, not only for melanoma and for those 20% of patients who don’t respond to treatment, but also for other areas where it’s been very difficult, like pancreatic cancer and others where immunotherapy doesn’t work. We want to learn a lot about this technology, while also using what the industry has learned in the last 10 years, because it has learned a lot, as you know.

This isn’t even intismeran autogene 1.0. That’s why I’m so excited about what lies ahead.

This is fantastic. Looking back, you and I have known each other for a very long time. I won’t disappoint you or myself by saying how long. You were in kindergarten, so I had the advantage of watching Moderna from its very early days.

7. The Operational Lift of Making a Vaccine for Every Patient

One thing remains the same from then until today: You are, above all, an engineer at heart. You were obsessed with processes, operations, and efficiency from the beginning. These things have to be absolutely flawless if you’re going to do what you’re trying to do here with personalized cancer vaccines and make drugs for each individual patient, precisely because 90% of the time the antigens don’t match.

Can you tell us a little bit about the operational scale that is needed here, that you have already secured for testing, and that you will have to secure if you eventually start commercializing this product?

Of course.

It’s probably worth starting with what many people think of as another big personalized therapy that exists in cancer treatment: CAR T-cell therapy. Isn’t that right? In this case, the essence of CAR T-cell therapy, for those who may not be familiar with it, is the idea that you take the patient’s tumor and then take the patient’s immune cells. You take them out of the body, reprogram the immune cells, reengineer them to respond to the tumor, and put them back into the patient.

I’m simplifying things, of course, but that is CAR T-cell therapy in short and in general terms. In this case, you’re doing very similar things to some extent, aren’t you? You take a piece of the tumor that you probably got as a biopsy, and you try to sequence the tumor to create a vaccine that is very specific to that patient’s tumor.

8. Vein to Vein: How Fast Can You Deliver a Personalized Vaccine?

What do you think about vein-to-vein time, essentially? What needs to happen from the moment you examine a patient and gain access to the tumor to the moment that patient receives their personalized vaccine?

Now that time is about 42 days, from needle to needle—that is, from taking a biopsy to receiving the finished vaccine at the hospital. I think we can improve this metric, as we still have a lot to work on in terms of efficiency, automation, and robotics.

The difference between this and CAR T is that we don’t have to take your immune cells. We program them outside the body in a reactor at our factory and send them back to your hospital.

The only thing we need is information. As we’ve discussed, the great thing about mRNA is that it’s an information molecule. We get the sequence of your healthy cells and the sequence of your cancer cells from the lab. We just get the file.

Then we use that information to actually create DNA. But now we don’t do it with plasmids, growing E. coli, or anything like that. We do everything synthetically. It’s all enzymatic in an aqueous solution. Then we create RNA based on the template. Then we wrap it in lipid.

Because it’s a synthetic process, it’s more like a small molecule than a large molecule. If you think about CAR T, for me it’s an analogy to the recombinant world, where you have cells, large reactors, and large volumes.

Because, as you know, the reason for the high volumes in the biotechnology industry is that if you squeeze cells too much, they will die. The same problem exists with CAR-T therapy, so all this requires a large scale. Here, however, since everything happens in an aqueous environment and with the help of enzymes—that is, it is very catalytic—the reactors are very, very small.

So, what we were doing before the clinical trials—because, as you said, we had to start developing the technology to individualize it for each person just to do a phase 1 study—was reduce all of that. The first version of the machine, which looked like a large American refrigerator, was massive and clunky because we told the team, “Make it good enough, but we’re not going to optimize the efficiency yet, because if it doesn’t work in the clinic, what’s the point of spending 5 years building a great, amazing, optimized robot if the science doesn’t work?”

So we said to the team, “Do it well so that there are no quality issues and we don’t get a false experiment, a false negative in the clinic, because it would be terrible for the patients if the robot didn’t work somehow.” You do a study, it shows that the science doesn’t work, and that’s it, even though it should work, right? This would be a disaster for humanity, of course.

And the team did just that. They developed a robot that was of good quality but not very efficient. When we got the phase 2 data showing that it was working—the first interim phase 2 data in just 2 years, and now we have 5 years of data that really supports the duration of effectiveness—we said to the team, “Okay, now we’re behind.”

This was supposed to happen in the case of success, which is a nice problem. So we brought in a lot of very smart engineers to think, “Okay, now how do we make a very efficient machine so that we can even reduce the size of the machine itself?” Because one of the important factors, of course, is time, as you mentioned—the cycle from needle to needle—but also cost.

One way to reduce the cost is to reduce the area it occupies. Because if you have a fixed cleanroom space, you can put, say, 2 or 10 times more machines in the same area. Of course, you will get more throughput and a much lower price for fixed costs.

So we are fixated on cycle time because the more we can reduce it, the faster we can get asset turns per year. The second vector that I’m focused on is square inches. I’m constantly, excuse me, annoying the team when I come to the factory to see how much space we can save, how to be creative, and how we can move some of the computing power out of the clean rooms to compress everything as much as possible, because ultimately this greatly affects the cost of the product.

9. Manufacturing at Scale & the COGS Question

And on a scale, roughly—let’s say we focus on melanoma—how many doses need to be produced per year?

We already have thousands of these doses through 9 ongoing clinical trials. A facility in Marlborough, Massachusetts, will be able to produce tens of thousands of such doses. As the technology improves, this number at the same facility will grow, and then we might have to build a few more factories.

But if you look at the incidence rate of melanoma, with tens of thousands of doses, you could easily cover the entire melanoma market.

Yes, I can believe that. Have you revealed what you think about the cost and price? Is this something that hasn’t been revealed yet?

We first need to disclose the data to our colleagues. We need to attract payers. When will we be able to share data with them on what value is being created, et cetera? But this has not been discussed yet.

Perhaps it’s worth focusing on the concept of personalization. I’m sure you’ve seen the story of the founder of GitLab who went into “founder mode” regarding his own osteosarcoma. First, will future “Sids” of this world come to Moderna, or do you think this “N=1” phenomenon will simply happen in parallel?

I think most of them will come to Moderna because it will simply be easier and safer. As you know, the production of an injectable product always carries a risk of contamination. If you give someone a drug that has even 1 bacterium in it, you can cause the patient to develop sepsis.

Then there is always the question of quality, because when the process consists of many stages, an error can occur. And of course, if it is industrial and meets GMP standards, such as FDA requirements, the likelihood of this is much lower.

It’s like with any tool in life: do you make your first knife because there are no shops, you’re in a cave, and you need to feed your family? Yes, of course you’ll make your first knife because you have to feed your family, right? But when there’s a shop around the corner with quality knives, you’ll spend your time doing something else.

So I think it’s the same phenomenon as in any technology: when there’s an industrial scale of a quality product, you, as a person, spend your time on something else, right?

10. How Do You Regulate a Medicine That's Different Every Time?

And in this world, how does the regulatory environment—the regulatory apparatus—work? In other words, you mentioned earlier that you will be preparing a regulatory application with Merck soon. What exactly is subject to regulation here? Since every dose is different, it’s obvious that not every dose passes approval. Is it the process of mRNA synthesis, or is it an algorithm? Is it a combination of the whole system? Help us understand this and form an idea of how we think such personalized medicines will be regulated in the future.

Yes. The good news is that there are precedents. As you mentioned, CAR-T was also approved in the same way we believe our technology will be approved—as a process BLA, and not as a separate product BLA.

As you know, Moderna has 5 approved products. So, 3 approved products. In this case, the entire process from the start of clinical trials to the IND was a “process” IND. Because we had to ask the FDA: Can we do clinical trials? Do you think this is safe? And do you think we have good control over the process so we can safely conduct a phase 1 study?

We already had this discussion before the start of clinical trials many years ago. Then, before starting each phase 3, we need to have a meeting at the end of phase 2 and agree on the study design and manufacturing protocol with the FDA. These discussions have been going on for years.

It’s not like we haven’t been talking to the FDA for the last 10 years, and then suddenly, in a week or 2, we show up on their doorstep and say, “This is a new product,” and they’re like, “What is this?” There were a lot of discussions and a lot of interactions.

We had a few technical questions about manufacturing, where we asked for additional meetings to get advice and also to tell them about the technology, what we learned, et cetera. So there’s been a lot of discussion already, and there’s a very clear regulatory path to get the whole process approved.

Basically, Jorge, what the FDA wants to know—and it’s a very valid thing, and I would want that for my own family—is: If you take the same sample at the beginning, do you get the same product out of the black box? That’s something we have to prove to ourselves first and then with the data that we have, so that we have real robustness of the whole process.

If we have the same input, we’re going to get the same outcome for the patient as a personalized medicine.

11. Beyond Melanoma: Which Cancers Are Next?

How do you plan to go beyond melanoma? Is that an approach that would be applicable to a broad range of cancers? Are there cancers where this is going to be a much more likely option than others? And what are your ambitions for where cancer vaccines can have an impact?

They’re probably high, because we think we’ve proven it at least to ourselves, and hopefully to the world. There are always going to be skeptics, but that’s only natural.

We’re able to create a new population of T cells, and we even demonstrated it at ASCO this year, taking blood from melanoma patients before and after treatment with our technology. We showed that not only are T cells expanding, but new, de novo T cells are emerging that recognize something we’ve encoded in the mRNA that wasn’t there in the patient before treatment.

To me, we’ve already proven to ourselves and to the clinical community that Moderna’s mRNA technology can teach the immune system to create new T cells to attack cancer.

Based on that, we’re working on 3 different avenues to expand applications outside of melanoma. This study, as a reminder, involved patients with stage 2, 3, and 4 cancer. They are the ones who were involved in this phase 3 study.

We are working everywhere Keytruda works. As I told you, we think the mechanisms of action of PD-1 and Moderna’s individualized mRNA are completely orthogonal. We think that allows us to have a synergistic effect and improve efficacy for our patients.

We are in phase 3 studies in lung cancer. We are in phase 2 for kidney cancer and bladder cancer. So we have a whole series of studies going on where the world knows that Keytruda works because it is already approved there. We believe that you will see a significant improvement compared to using Keytruda alone.

Until you do a clinical trial, you cannot know whether you will get 50%, as we saw in phase 2 melanoma, or 30%, 40%, or whatever. We have to do the research. So a lot of things are in the pipeline right now.

The second vector is early-stage disease, where checkpoint inhibitors work. The best example is that we announced in the spring of 2026 that we’re going to start a phase 3 study for stage 1 lung cancer patients, but as a Moderna product, as a monotherapy, without checkpoint inhibitors.

Mhm.

And we’re doing that because we think that in early-stage disease, inhibitors are not used because of the side effects that they cause. Because if you have stage 1 cancer, medicine thinks it’s better to watch the cancer than to give an inhibitor, because not everyone is going to respond to it.

But everyone is going to have some pretty serious lifelong side effects, like autoimmune diseases. What if you could create an mRNA for stage 1 lung cancer patients that is easy to detect on an X-ray, for example, in former smokers—that’s the easiest target group? Just check your former smokers on an X-ray regularly.

And if you do that regularly, you go from a situation where you don’t see the cancer until you see it. The idea is to do surgery, which is the standard of care today, and then do a T-cell immunotherapy, which has side effects that are similar to a vaccine. You might feel tired one day, but that’s it. So that’s a pretty good type of side effect for cancer, right? And that’s another approach in our clinical trials where checkpoint inhibitors aren’t available today.

The third vector is where checkpoint inhibitors don’t work. So, of course, that’s where you’re at the highest risk.

Mm-hmm.

But because the mechanism of action is different from checkpoint inhibitors, we and Merck believe that there’s a very strong scientific rationale to try it. So the 2 areas that we’re testing right now are pancreatic cancer and gastric cancer. For those 2 types of cancer, checkpoint inhibitors like Keytruda don’t work. Clinical trials have been done in the past, and the results have been negative.

But we think that, again, the mechanism of action is different from checkpoint inhibitors. And now that we’ve proven that we can make T cells de novo, we think that this experiment is worth doing. If we get a good signal, we’ll think about combining it.

As you know, just yesterday, Revolution Medicines got approval for a great new drug for pancreatic cancer with a KRAS mutation. What if you could combine these drugs with intismeran? These are very orthogonal mechanisms of action. It doesn’t make scientific sense to me to suggest that if intismeran works, it shouldn’t work in pancreatic cancer on its own.

And, of course, Revolution Medicines’ drug significantly improves survival in pancreatic cancer. If you combine those 2 things, we believe that there should be an advantage. Again, we need to do a clinical trial to see how much, but those are the kinds of things that we want to do.

So if we look at intismeran as a monotherapy, it would be used with Keytruda. It will be used in the early stages of the disease with Keytruda, and also where Keytruda doesn’t work but with other agents.

Well, it gives a good reason, I think, for cancer patients and their families to have a lot of hope for the future of the new treatments in this area.

Yes. And in addition to what we just said, remember that this is intismeran 1.0. What I saw is very powerful, and I’m really pushing our team to think outside the box, to do a lot of analytics, and to use AI to study the huge data sets that we have.

The question is: What can we learn from clinical trials to understand people who haven’t responded to treatment? Because I think you always learn more from what doesn’t work than from what does work. So I want to focus on those 20% of patients who don’t show a response after 5 years, so that we can understand why they haven’t responded. Can we adjust something in the algorithm or technology to help them?

12. The Full Arc: From COVID to Cancer to Autoimmune Disease

Well, that’s extraordinary. And finally, I think it’s just amazing how you’ve been able to take a technology platform that wasn’t originally designed for a pandemic and turn it into fighting a pandemic, creating a vaccine for millions of people, and essentially, 10 years later, as you described, going back to treating the diseases that it was originally designed for. And that’s going to be really exciting.

Maybe by the end of the year, we’ll have a pivotal or late-stage study in rare genetic diseases in children who have rare genetic liver diseases. So that’s another direction that we’re taking the technology. In phase 1 and phase 2 of the study, the kids have been on the drug for 3 years, and they’re doing great. So we’ll see when we get that data.

And in June, we had our annual Science Day, where we announced that the next frontier that we’re taking our mRNA platform to is autoimmune diseases. Because if you think about it, we’ve learned a lot from infectious diseases—diseases that are mediated by the immune system. We’ve just talked a lot about cancer in the context of the immune system.

We’ve learned so much about the immune system that we think we have some very novel approaches to treating the root cause of autoimmune diseases, rather than the symptoms, which is what the pharmaceutical industry is typically doing. Of course, treating the symptoms is very helpful for patients to improve their quality of life, but it doesn’t address the root cause.

And we think we may have found ways to use the immune system to treat the root cause of autoimmune diseases. So there are many more ways to use this platform. So we’re very excited about what’s ahead.

Is the theory that you’ll have personalized immune modulators? Or is it more of a product or a process?

We’re working on both. What we introduced in the spring is a product that would be the same for everyone. But what I’m most excited about is what’s still in the lab: the possibility of personalized autoimmune treatments where you go directly at the immune cells that attack your body as if it were foreign when you have an autoimmune disease.

You’re essentially forcing a part of your immune system to attack those immune cells that are not working properly in order to eliminate the symptoms of the autoimmune disease. Again, this is just the beginning, but this is what I’m most excited about today.

So, moving from infectious diseases to cancer and then autoimmune diseases, we’d love to have you back on the podcast to record a 3rd episode and complete the trilogy when you’re ready.

Great.

Stéphane, thank you so much for joining us. As always, it’s great to see you.

You’re welcome.