We're going to get started with our next session. I'm Andrew Tsai, Senior Biotech Analyst at Jefferies. Thank you for joining me today. I have the pleasure to have the Moderna team with me. To my direct left is David Berman, Chief Development Officer, and to his left, Lavina Talukdar, Head of Investor Relations. Welcome, both of you.
Thank you for having us, Andrew.
Sure. Congratulations on the success that you've had so far with that phase III INT melanoma data that was just shared at ESMO. As we think ahead into 2027, can you just walk us through the various milestones you would encourage investors to pay attention to? That'd be helpful.
Yeah, Andrew, once again, thank you for inviting us here today. For the rest of the year, I think there's two more data points. First is our mRNA-2808 T-cell engager myeloma data, which will be at ASH next month, and then a readout of our propionic acidemia pivotal trial. This is our first proof of concept that we can develop an mRNA LNP for a metabolic disorder. Next year, or either towards the end of this year or into next year, so sometime in the next 12 or so months, would be the RCC interim analysis data, and then MIBC data final analysis would be in 2027 also.
Both for intismeran.
Yeah, for intismeran.
For intismeran.
Correct.
Would lung cancer, by chance, since you complete enrollment for one of them, be ready year-end of 2027 possibly?
It is event-driven. In theory, it could, but it could also be in 2028. Either 2027 or 2028.
Okay, very clear. I did want to spend a lot of the discussion on INT today, if you don't mind. Big picture, can you just remind us how big the TAM is for INT, starting with adjuvant melanoma, where you've succeeded in phase III, and then walk us through the next set of TAM? That would be very helpful, I think.
We do believe that intismeran is going to be a multi-billion dollar opportunity. But for now, we've given the eligible population for adjuvant melanoma, because that's where we have the positive phase III study. For that indication alone in the U.S., it's 22,000 patients that are eligible based on our phase III clinical trial inclusion criteria.
RCC, bladder, lung, how big would those be?
Also multi-billion dollar opportunities. We have not yet given the eligible populations yet.
Okay, no problem. Looking ahead, as we think about commercialization, because you're filing the BLA by year-end, this could be approved in second half 2027. Speak to the commercial readiness. How much have you prepared? Are you ready to go on day one, should this be approved?
I'll start, and David, please fill in. You're right that we are advancing the BLA and hope to get it fully filed by the end of this year. We are commercial-ready. Just last week, we hosted investors as well as analysts at our Marlborough facility, which is purpose-built for the intismeran product. It is not fully built out yet. We are using one suite, which is enough to get through commercialization for the melanoma setting, and we do plan to add on up to six additional suites as we see demand signals, and that, to us, is positive readouts from the very large development program that we have ongoing.
I'll just add a couple other points, Andrew. From a launch perspective, I think, if you think about the drug profile, great efficacy and the adverse event profile is very mild. It's essentially just some fever, flu-like feelings. From a patient-doctor experience on treatment, I think it should be very manageable and very favorable. I think for us, what we're focused on is the front-end part, which is teaching sites about obtaining the right amount of tissue and type of tissue, and getting that then submitted for sequencing and for the drug product to be made. The good news is that we've been able to do this, obviously in a randomized phase III, but now for thousands of patients globally. We are optimistic that we will be able to serve the launch appropriately.
Right. Through your clinical studies, you have already dosed 1,000+ patients in-
More than 2,000 patients.
Okay.
When we are fully done with the clinical development program, we will have enrolled 5,000 patients. Your question on being launch-ready day one, we feel very optimistic that we are launch-ready.
Okay. Going back to the Marlborough site, I think you mentioned one suite is ready now. How much capacity is that? How many patients per year can that supply? You said six additional, so what's the total amount per year off of that single manufacturing site?
When we're fully built out, we'll be able to supply tens of thousands of patients from that one facility, but that's at full build-out. We haven't said anything about the capacity from the one suite, as we're continuing to do productivity and efficiency gains on the process itself. It is a continual process that's ongoing.
I see. Understood that this alone can probably handle globally, but are there other manufacturing sites that you've built out or are building out globally?
This facility, we do think will have global reach. Just to remind everyone, the clinical trial program in adjuvant melanoma, as well as some of the other programs that are ongoing, has served 46 countries around the world, and we did that without a hitch. We're very optimistic that this Marlborough facility on the East Coast of the United States can actually do a global launch.
Speaking of efficiencies, I think one discussion point could be the turnaround time of INT. In the clinical studies, remind me, how long that took. Where do you think that can get you to over time?
In the clinical trials, it was about 63 days. Part of that was also because in the clinical trials, the intismeran or placebo had to be matched to the dates of the pembrolizumab infusion, which doesn't need to be the case. We know we've been able to get it down in a good situation to about 30 days, and I think about six weeks is something right now that is very achievable. We will continue to bring down that turnaround time.
I see. In terms of the manufacturing success rate, also remind us what that was in the clinical studies. Is there any need to be improved?
Yeah. We have great success rate. Once we have adequate tumor, we have a very high success rate of identifying neoantigens. From a manufacturing point of view, it was over 99% successful. In fact, I think only one of the 700+ patients in the intismeran randomized arm was not able to receive their drug product. So, a very high, over 99% success rate of manufacturing.
I see. Can you remind me the Merck partnership? This is a collaboration. To be clear, you're building out the drug supply. Is Merck selling it? How does that work, U.S. and ex-U.S.?
The partnership with Merck is truly a 50/50 profit share partnership. Merck and Moderna will co-market the product in the United States, and Merck is fully responsible for commercialization outside of the United States.
I see. Ex-U.S. What other countries are you and Merck filing to next, and when does that happen?
I don't think we've disclosed where the global filing will be. Right now, we're focused on the U.S.
Got it. I asked you last week, but I will ask again, is just, we need to shore up our models on how this is going to be booked on the Moderna line. How do you recognize this on the P&L?
We are still discussing that with our auditors as well as Merck and their auditors. There is a scenario where we, because we manufacture the product, that we do pre-booking in terms of the sales that will be part of our revenue line. However, that is still being worked out with our auditors.
I see. Revenue line. Okay. Do you book in terms of revenue, INT only revenue or INT plus PD-1 revenue?
It is INT only.
Okay. Gross margins, COGS, how should we think about that initially? Some people wonder if this is going to be like a cell therapy launch, high COGS, low margins. Can you walk us through how you're thinking about that and where does that go at steady state once you scale?
Over the long term at steady state, we expect to have very attractive margins for this product. We do not anticipate this looking like CAR-T, either at launch or over time. Definitely not.
Got it. Pricing bookends for INT, what should be some of the analogs we can gravitate or look at?
A little premature to talk about pricing. Typically what our partner Merck has done in the oncology setting is that we'll set pricing upon launch. Just to help on modeling, a very large range, but you can think of our proposition of a value proposition in terms of pricing, and the range maybe you could think through is low-end KEYTRUDA-like pricing and high-end other analogs in the market include individualized therapies like cell therapy. That's a very broad range. Somewhere in between that range could be a reasonable estimate.
Thank you. As for the application itself, because you're using this kind of proprietary unique AI algorithm, does that require some kind of special approval on top of the drug side? Maybe talk to that.
Yeah. The algorithm, the proprietary algorithm is part of the drug manufacturer, so it does need to have FDA review as part of the CMC approval.
Okay. As we think about you building out a moat for INT, can you talk about why other AI competitors won't be able to catch up or beat you guys?
The most important thing in an algorithm is linking it to efficacy. That's the single most important thing. There are a lot of algorithms out there, and they can generate T-cell reactivity, but you never know if it's going to be efficacious or not. Number one, we know that our current algorithm works, and that algorithm is proprietary. Number two is we now, with efficacy data in hand, we can go build a version 2.0. We are using AI as well. But we can build a version 2.0 by linking back to what makes a quality neoantigen, not just any neoantigen. I think that we currently have a large moat, and I think we're going to, with so many ongoing randomized phase II and phase III, nine of them, we're going to build up the world's largest database on what makes an effective neoantigen.
I think it's going to be hard to overcome.
Yeah, and that data really needs some emphasis because it is proprietary data that we will have that will then inform the algorithm version 2.0, version 3.0. From a competitive standpoint, it's that data that's going to inform the selection of the neoantigens. The other important thing to consider here in terms of a moat is that we took the strategy of starting in the adjuvant setting, and the studies that we're running is against the standard of care. In melanoma, for instance, it's against pembrolizumab or KEYTRUDA. If we do then become the next standard of care, the combination of intismeran plus KEYTRUDA, these studies in the adjuvant setting can take multiple years to read out. A competitor coming to market would have to use us as the control arm, which can also be cumbersome.
I see. And then maybe shifting gears to the next readout, RCC year-end 2026 or early 2027. Are you guys using the same algorithm for that one, as well as for the other programs, actually?
Yeah, we are using the same algorithm. It is the algorithm that works for melanoma, and it should work for other tumors as well. So it is the same algorithm.
I see. To the extent you are willing to share, as we think about the scenarios of the interim, if it does not hit stat sig this time, does the study continue, or is there a futility?
Yeah, the study will continue. It is an interim for efficacy, not for futility.
Got it. Obviously, if it hits stat sig, I think you will announce that for sure immediately. If it does not, do you announce that immediately?
This is an interim on a phase II, and I think we're in discussions with Merck on exactly how would we announce it, because it is a phase II interim, it's not a phase III interim.
Okay. I think you guys have done a good job elucidating why you feel confident in RCC, but can you just remind us all the pieces of evidence that has led you to why you feel confident?
Yeah. Sure, Andrew. First of all, RCC is a T-cell-sensitive tumor. We know cytokines work, and we know that immunotherapy works here. Our mechanism is to generate novel T- cells, checkpoints to reactivate T- cells. That's what we've learned from melanoma. That's point number one. We can generate novel T- cells that the patient doesn't even have. Number two is, there's a very nice paper out from Dana-Farber looking at personalized cancer vaccine using peptides, using their own algorithm. On average, in using their algorithm, they identify 45 quality neoantigens. I think that's a good surrogate. We have a different algorithm, and we haven't disclosed how many we identified, but I think the fact that it's above 34, which is what goes into our cassette, I think that's a hopeful sign.
Number three is our tumor mutational burden analysis from our randomized phase II. We had a similar efficacy in the phase II, whether it was high or low TMB, and that low TMB bracket overlaps with RCC. Finally, the thing that was most compelling to me is the treatment effect that we saw in our phase III melanoma trial. Here, it was 0.58 hazard ratio, so 42% improvement, which is essentially what KEYNOTE-054 showed in the adjuvant melanoma trial. Then, of course, we know that pembro went on to work in RCC, bladder, lung, et cetera. We feel there's a lot of headroom left, given our benefit. I'll add, our top line, our ITT, of course, was 0.58 hazard ratio, but our modified ITT, that is patients who actually received the blinded vaccine or the control, that hazard ratio is 0.51.
That, to me, represents the true treatment effect. This is a really powerful treatment effect in melanoma. Those are the reasons I'm optimistic about RCC.
Oh, great. Then you mentioned just now the TMB in phase II as one piece of evidence. Have you done that analysis in phase III, as well as ctDNA, BRAF subgroups, and what are you seeing?
Yes, so those were all the interesting insights we made in phase II, and they are now in process for the phase III trial, and we'll share those data when they're ready.
Okay. For RCC, what is the minimum number of months of follow-up these patients will have?
Well, I can tell you now that we finished enrolling that sometime last year, I think in the summer.
Second quarter.
Yeah. So, it has already been over 13, 14 months. So, it is going to be long enough to go past whatever theoretical delayed separation that could happen. I think that's hopeful. The minimum follow-up will be whenever the events get analyzed.
Thank you. Do you guys have an internal bar with Merck, what kind of efficacy you want to see? This is a combo versus PD-1. I think there's HIF-2 combo versus PD-1. Is that the bar, or not necessarily?
Yeah, not necessarily. Just as a reminder, belzutifan plus pembro had a hazard ratio around, I think, 0.72 or 0.73. It does come with some HIF-2-specific adverse events, like hypoxia and anemia. Our intismeran comes with very manageable, mostly mild adverse events, as I mentioned, of injection site pain and some mild fever. So, I think there is that efficacy out there, but we also have a very favorable safety profile compared to belzutifan.
Yeah. I was just thinking in my head as we're discussing, the original phase I, there was a basket study, but I noticed, I don't believe you guys tested in RCC. The question would be, why not at the time?
Yeah, RCC was not part of the original basket trial. It included, I think, the typical tumors, the bladder, lung, and melanoma trials that historically have been very immunosensitive. But I wasn't here 10 years ago to know why RCC wasn't included. But I think based on what I know now, I am optimistic this can work in RCC.
Yes, understood. Okay, then moving on to the second INT readout coming up, bladder. Remind us the standard of care. I think this is also combo versus PD-1 alone. What is that efficacy threshold that you want to see ultimately?
Yeah, so MIBC, just as a reminder of the trial design, it's randomizing to intismeran plus pembrolizumab versus pembrolizumab after removal of the bladder. Now, here, the standard of care has changed since the trial started because now you have EV plus pembro in the neoadjuvant setting. In fact, by the way, we do have a cohort of EV plus pembro plus intismeran as a just trial combination, neoadjuvant, adjuvant, because we anticipated with Merck that EV might become the standard of care. So we do have a separate cohort. It's not part of the randomized trial. EV plus pembro was not included, obviously, in our trial design, so that standard of care has changed.
You also have atezolizumab in ctDNA positive after resection, that was not obviously included in our trial, although we did stratify by ctDNA because we did anticipate that ctDNA would be important. I think that there has been changes since the original trial was designed, and I think we'll look at the treatment effect that comes out of here to determine what exactly the next phase III study would look like.
I see. So there could be a chance that you ultimately run a head-to-head type study, or not necessarily?
I think that there are multiple options because of the changing landscape. If you look at what other companies have looked at, we could consider a triplet neoadjuvant, so EV plus pembro. I think EV plus pembro needs to be the backbone in the neoadjuvant just based on the data. But then there's a question in patients who receive EV plus pembro and do not have a pathologic complete response. There might be a chance for intismeran in non-pathologic complete response. I think that more needs to We'll need to see the treatment effect size and what the landscape looks like next year.
Yeah. Then, I think we discussed earlier, lung could be year-end 2027, 2028. Remind us what exactly is this subgroup for lung? Is it the INTerpath-002 that just-
Yeah.
in enrollment?
Yeah, correct.
That's a big study, right?
Yes, it's a big study.
Phase III.
It's a randomized phase III. It's in patients. We have three phase III trials in lung adjuvant. We have the INTerpath-002, which is patients who never received any neoadjuvant therapy, randomized to pembro plus intismeran versus pembro. That study completed randomization this year. Very fast enrolling trial. That's the one that events could happen at the end of next year, 2027, or into 2028. We have another trial in the adjuvant setting, patients who receive neoadjuvant pembro plus chemo but did not have a pathologic complete response, which is most patients in that setting. That trial is still enrolling. Then a third one that I'm really excited about, which is in stage one lung cancer. This is the earliest where, for the most part, the standard of care is observation. You wait and watch until the patient relapses. Here, we're testing intismeran monotherapy as one of the experimental arms.
I see. Okay. I think as we continue to think about the various pieces of evidence, back at ESMO, you guys did actually have a pancreatic readout. I think it is INT plus FOLFIRINOX. What did you see there, and should that have read across to these upcoming INT readouts?
Number one is, we were able to manufacture cassettes. In over half the patients, there was a full 34 neoantigens put in, which I think is pretty remarkable. Every patient had a T-cell responding to at least one neoantigen, and you could combine it with FOLFIRINOX, and after FOLFIRINOX ended, we still had higher T-cell reactivity than compared to baseline. This can be studied in adjuvant. It was just too small, and the follow-up is not long enough to know if there is any efficacy signal.
I see. One more is just the competitor front, for instance, I guess I get asked on BioNTech, and so why do you think you are the superior personalized vaccine?
There is a Right. I mean, I do not want to say superior.
Okay.
I think we're the one that has demonstrated efficacy in a randomized phase III trial. I think there's some scientific platform differences. We use 34 neoantigens, they use 20. We use the same LNP that we use for COVID, injected into the muscle, and we know it goes into the draining lymph node. They inject it intravenously, and it goes to the spleen. We don't know what their algorithm is, but we know that our algorithm has produced an efficacy benefit, and our turnaround time has been faster than their reported turnaround time. I think that those are the differences, but we have the randomized trial.
Right. I think you guys have taken the data-led approach because in the phase I, you did test colorectal already, and you didn't want to choose to enroll.
I still think colorectal is something that needs to be explored at the right time.
Okay. Yep. To tie everything together, you're really, to me, I feel like going all in on INT. I think you've once guided 2028 to break even. Is that still the case by chance? No worries.
We just had our Analyst Day, as you know, Andrew, and what we said at the Analyst Day is that our infectious disease vaccines portfolio, that business is on its way to profitability, and we expect that business to fund and sustainably fund itself. Any new infectious disease programs that will start will be funded by the cash flows in that business. But with this remarkable result in intismeran, we do now have a balance sheet that is $9 billion strong at the end of this year that will fund the growth in oncology. The continual progress in the program, that development program we have ongoing, and any new programs or development that we do there.
Perfect. I think that's all the time we had, but thank you for-
Thanks, Andrew.