第195期 - 2026年9月11日

You're listening to Biotech Hangout, a live and unedited weekly discussion of all the latest news in our industry with a group of biotech experts and leaders. I'm Tess Cameron, and my co-hosts today are Brian Skorney, Paul Matteis, and Yaron Werber. For more information about our hosts and guest speakers or to listen to the most recent episode, please go to biotechhangout.com. Thanks so much for joining, everyone. We've had an interesting week. It's been a somewhat challenging back-to-school week for the XBI, which is down just over 3% over the past week or so. It's been a pretty challenging tape overall.

There was a significant sell-off in the stock market overall, due to oil prices spiking in the U.S.–Iran war, with Brent crude getting above $108 a barrel. Ten-year Treasuries reached highs they haven't seen in many years—4.8%, and then close to 5% on Thursday. A lot of that was linked to concerns about energy prices, the state of inflation, and the long-term fiscal health of the U.S. economy.

1. Rates Return To Biotech

That obviously has an impact on biotech. Biotech has always been a rate-sensitive sector. We saw biotech decline a bit more than the overall S&P this week, but not as much as the tech-heavy Nasdaq index.

I'm interested in any comments from Brian, Paul, or Yaron. A couple of years ago, we were living in an environment where biotech and rates were very closely linked. Then 2024 was very much a policy-driven year. Are we getting back to rates really being one of the driving macro, sector-level factors?

The way I've always characterized it—and maybe this morning, when XBI was up on a hot inflation number, I questioned whether that was right—is that it's not so much rate-sensitive to small up-and-down movements. It's more about whether we're going to see rates escalate in 100-basis-point moves over a year or two.

When you see the federal funds rate going to the high single digits, that's very bad. Going toward zero is very good. But when you're talking about a 25-basis-point change here or there every now and then, you could deal more with fundamentals.

I guess it's not surprising that we're talking more and more about inflation being a theme, and the Fed potentially escalating rates even though Trump has been such a big advocate for lowering rates. His appointed chairman seems to be a little more hawkish than perhaps he was in the Senate confirmation hearings, and maybe that's a little bit of a break with Trump himself.

I think there is that theme. Inflation is a funny thing, so we'll have to see how much it continues to drive. If we just see 1 or 2 rate hikes, I think this will be much ado about nothing.

2. FDA Leadership Calms Biotech

Thank you, Brian. Let's move on to some of the fundamentals. We can start with regulatory policy.

HHS appointments were announced earlier this week for directors of CDER and CBER. HHS announced that Karim is going to keep his place as director of CBER, making a temporary appointment permanent. It also announced that Michael Davis would be selected as director of CDER, also making that appointment permanent.

Those are the 2 main appointments that really have read-through to biotech. There were also appointments for the Center for Tobacco Products and a new appointment as deputy commissioner for technology and artificial intelligence, Jared Seehafer. Brian, could you share any views on these selections and what they mean for biotech?

Happy to. I would love to have input from the rest of the group here. Nothing against Jared and Brett, but they probably don't matter in the context of analyzing biopharma. The big names to look at are Michael and Karim.

Over the last, call it, close to a year, the sector as a whole has been breathing a little bit of relief as we've seen the chaos that came with the appointment of Marty Makary as head of the FDA, with Vinay Prasad as head of CBER, and then subsequently as CMO and CSO of the FDA. That's been a lot of chaos.

I was looking back through the history. The CDER and CBER roles have been largely apolitical in nature. They're usually, if not lifetime appointees, longtime bureaucratic appointments. Janet Woodcock served from 1994 through 2020 as head of CDER. Peter Marks didn't do it as long, but he was there from 2016 through 2025, so he must have had a 10-year stint.

Since the new administration, the CDER director has turned over 5 times. CBER leadership has been headlined by Vinay, and we all have a number of opinions on how Vinay ran things. The nicest thing we could say is that there was a lot of chaos and a lot of senior leadership departures.

There seemed to be a lot of discontent within the FDA, and that led to questions about agency guidance and decision-making processes. The industry was struggling with how to work with that FDA leadership.

Since the departures of Vinay and Marty, we've seen a calming. It's a relative calming. I think there are still a lot of questions about whether we've clearly seen the kind of leadership that we saw from Janet or Peter—not necessarily a good thing. I recognize that there are debates about their leadership as well, but compared with Vinay and Marty, there were a lot of issues with them.

I think the nicest thing to hear in the commentary that Karim has made since he was put in the interim position is that he wants to be more of a calming force. He says he wants to be like a conductor; he doesn't want to take over the violin when the violinist isn't doing well. His role is to be the leader, not necessarily the decision-maker.

I think those things speak to how most of the industry wants to see the FDA run. We want to see scientific discourse and groups working together to come to decisions about when a drug should be approved and when it shouldn't be approved.

This is all in the background of the nomination of Dr. Heidi Overton to the role of commissioner of the FDA. It remains to be seen whether she'll be able to secure the needed votes, but she appears to be more of a stabilizing force than her predecessor.

These are all certainly steps in the right direction for the industry relative to what we had a year ago. There are still a lot of open questions because we haven't seen a ton of direct actions taken by the people who are now in these positions. At least from what has been said, the general feeling in the FDA seems to be one of much calmer seas than a year ago.

Let me just add that I have to give credit to Mike McMahon, who's one of the founders of Provision Policy and was a dear colleague of ours. Sadly, he wasn't able to join, but I'll pass along some comments.

He believes that the decision to stick with Dr. Davis and Dr. Karim is a pretty good outcome. You can think of many ways this could have gone worse, and not many ways it could have gone better.

Specifically, the decision to stick with Karim for CBER is a good outcome. He's a little bit of an odd fit for CBER director, but given how erratically and poorly Prasad did in that role, McMahon thinks that, over the short term, Karim will do a much better job stabilizing things.

So, the bottom line is that he felt, net-net, it was more of a positive than not.

Yeah. Yep, I’d reiterate that. Michael has been a part of several listening sessions and other points at the FDA, and I think is acutely aware of a lot of the organizational challenges that CBER experienced under Vinay and very sensitive to them. So, as someone who was a former operator and had to build and lead large teams, I find it very reassuring to have someone in that role who has a real appreciation for how to do that effectively and how to be a good leader despite maybe having a different background than one may typically have expected for CBER.

Maybe we can also comment on some of the stock reactions to this appointment, and there are a few areas that Davis in particular has been known for. One is psychedelics, which he’s made some comments about, and another is compounded GLP-1 products. So maybe, if someone wants to comment on the psychedelics—his psychedelics commentary and impact on stocks—I can take the GLP-1 point.

I can do that. Can you guys hear me?

Sure can.

Okay. Hey, what’s up, everybody? I think we’ve talked about it a lot on this podcast, so I can keep it brief. But it does continue to feel like the FDA’s got a door wide open for psychedelic drugs. I mean, the FDA just put out finalized guidance for drug development, which, generally speaking, aligns with a number of the clinical programs that we’ve seen out there from Compass, Definium, and others who are in mid- to late-stage development. That’s as it relates to 12-week primary endpoints, redosing, and best efforts to mitigate functional unblinding.

There was also a publication. Was it in NEJM, or was it another medical journal? It came out in the past few days, came from Dr. Davis, and just reiterates the same sort of perspective. So, I mean, it’s an interesting time, right? Because Compass is about to complete its NDA in the fourth quarter. It’s a rolling review.

I think it’s also interesting because, on the investor side, we’ve still been getting questions about whether there could be AdComs here and whether there could still be lingering questions around things like functional unblinding. For Definium, there’s been this conversation around their next trial, and one of the doses is a lower dose: How subtherapeutic is it going to be, and how much does that matter? I think, big picture, clearly, if not already, we have leadership at the FDA that appreciates the unmet need in psychiatry and is very enthusiastic about this class of drugs.

Great. And I will take another topic that Michael Davis has been known to comment on, which is compounded GLP-1s. He’s made commentary previously on the importance of really understanding what patients are taking and having patients understand what they’re taking. He was behind, I think, several of the warning letters that went out to some of the compounding groups. So that’ll be another area to watch—another very contentious area—and we’ll see if there are any changes on that front.

Maybe, to wrap up on some of the regulatory and policy topics, I’ll share some takeaways from a conference this week. There were a couple of different broker conferences this week. There was the Wells Fargo conference in Boston, and there was the Cantor conference in New York. I had the pleasure of going to the Cantor conference, and they were kind enough to host a panel on China that I thought I’d quickly share some of the takeaways from.

This was a really interesting panel where they included a few different perspectives. They had Simone, who’s a partner at ORI, which is a China biotech investment fund. They had John Crowley, who is the head of BIO. They had Ken Song, who’s back as a partner at Venrock and is doing a lot of work around China and other geographies, really expanding the firm’s footprint. They also had me on this panel.

It was a very interesting discussion where I think John Crowley really had the opportunity to highlight some of the continued policy discussions that are happening related to China. In particular, he discussed the possibility that some kind of guidance or understanding of policy actions may come forward over the next couple of months. He shared BIO’s position, which is that, while this is something the US really has to watch very closely, and while he and BIO have put forward a number of recommendations for how to improve US competitiveness, any kind of ban would probably not deliver the intended results in terms of actually strengthening US biotech.

So, much more to watch there, but it was a big upgrade from last year. At the Cantor conference last year, I was also on the biotech panel and the China biotech panel, and it was in this little room kind of on the side of the conference. This year was a much bigger panel in the ballroom, so it was definitely an upgrade in terms of the interest in that conversation.

Why don’t we head over to data? There were a number of data readouts over the weekend and early this week. Yaron, maybe I will flip it over to you to start with Novartis.

3. Lp(a) Fails The Outcome Test

Yeah, absolutely. So Novartis, along with Ionis, actually had data from the long-anticipated Lp(a) study. It’s the data from the HORIZON study, which was around 8,000 patients. Those of you who have popcorn, this is the time to crunch loudly, because this was highly anticipated, and the data—drumroll—was dropped on Friday, literally at 4:30 p.m. after the market closed, into the Labor Day weekend. Immediately, you knew this data was not positive.

Little did we know that Novartis was going to follow it up thereafter with data from Avidity, which also was not positive, and it led Novartis to be down a lot, about 14%. To go back to Ionis, Ionis was only down very incrementally because it was largely expected for the Lp(a) Phase 3 not to work. Ionis then—that’s Monday morning—and then Pharvaris, which I think Brian’s going to talk about, had their own Phase 3 data in hereditary angioedema. They’re a competitor to Ionis. So then Ionis went down from about 2% down to maybe 5% because that data actually looked very good.

To go back to Lp(a), because this is really important, recall that there has been a lot of epidemiological data looking at Lp(a) as a genetic modifier, a marker for risk regardless of LDL levels and, let’s say, blood pressure and things like that. It was always considered to be an independent potential risk modifier for outcomes, and this was the first study. The population that they enrolled ended up being the most well-controlled population ever in any study. I think average LDL was like 60 to 65, very well controlled on blood pressure, and the mid-teens or so were on SGLT2s.

We sort of knew over time that the data was not going to be amazing. I think everybody was hoping for a low-teens MACE benefit, just given that this was the final analysis in the study and it wasn’t stopped earlier. Bottom line, it looks like it was a completely negative study. We’d actually been working on a big piece heading into that, and we decided, funny enough, not to put it out because we realized all the historical data was in populations that had very high LDL, and the most recent data that literally just came out at the last cardiology meeting, in a more controlled population, suggested that Lp(a) might not be that important.

So this was a little surprising and a little disappointing, obviously, in many ways for the field. That drug is pretty good, pelacarsen. It reduces Lp(a) by 80%. The next drugs—olpasiran from Amgen and the Lilly version—reduce levels by 95% to 100%. So we’ll need to see whether that makes a difference. We’ll need to see how well-controlled those populations are going to be in these big Phase 3 studies.

This was also secondary prevention. Both Amgen and Lilly are not doing primary prevention, where, frankly, Lp(a) might be more of a risk modifier. But the bottom line is that the initial go did not look great. I think, Brian, you probably wanted to comment a little bit on maybe Amgen and Lilly as well.

Yeah. Just from the context that they both have targeted therapies that drastically reduce Lp(a), I think one of the things that was interesting about this data point is that, as Yaron said, people weren't very confident in it. What I think was noticeable about the Lp(a) story is not how bearish, but how ambivalent, the biotech mafia has been about this working out, and we'll talk about some of those reasons.

But I have found it to be an incredible talking point among generalists, certainly on the Amgen side. We saw Amgen lose 10% on Tuesday because of this blowup and the read-through there. I fielded a lot of questions on their drug from people who are interested in the thematics of this: This is potentially one-fifth of the total population that has genetically induced elevations in Lp(a).

So it fits into the theme of where people have fallen in love with the obesity side of things because the numbers are just so dramatic. So what is the read-through? You mentioned that Pelacarsen is sort of the least active of the drugs that are in development among the large-cap pharma companies—Amgen, Novartis, and Lilly.

But they do about 70% to 80%, depending on which study you're looking at, Lp(a) reduction, which is very dramatic when you contextualize that LDL reductions of 30% lead to significant, clinically meaningful hazard ratios. So the idea is: Is getting from 80% to 98% going to be a meaningful difference? Do you have to eliminate all Lp(a)?

As you said, there's a lot of conflicting literature, and I think one of the big issues with these studies is that we're just very good at controlling for cardiovascular events. When you're intervening with statins, ezetimibe, PCSK9s, and GLP-1s, all of which benefit people's cardiovascular health, is reducing or eliminating Lp(a) really going to move the needle? Or, even if it is a causative factor, do we have to look at much, much bigger studies because of the standard of care?

The one thing I would point to, and one of the reasons why I think people have been bearish on this of late, is that this study has taken a lot longer than people originally expected. Novartis has said that the pace of events has not been what they modeled. I would say this was something I think we've talked about before, but as a general thematic, when these outcome-based studies have rates that are mismodeled, my experience is generally that the study is going to fail.

If you don't understand what the rate of events is in a study, you're mispowering for the events. You don't really understand what the background meds are. I've almost never seen it where that's wound up benefiting the treatment arm, and that's the case here, right?

While there can be a defense around Amgen and Lilly for having better Lp(a) reductions, there are trial design issues, too. The inclusion or exclusion of stroke in the MACE endpoint could be important. The baseline Lp(a) could be important.

But Amgen has sort of had the same case: This study is not going as fast as they expected. The outcomes are not occurring at the rate that they thought they would. I think that mismodeling is an indication that something's not going right with the study, whether the drug is not reducing a causative factor or the placebo is just doing way better than expected.

We'll have to see. I sort of expect that Amgen data maybe to read out next year, but they're not really committing to anything. The Lilly data may be in 2029, but it's hard to look through this data and have a lot of confidence that pharmacological Lp(a) reductions are going to make a material impact on cardiovascular outcomes.

Hey, Brian, can I ask you a question about the other studies—either you or Yaron? One thing that came up in talking about this—because I cover it from the Ionis angle, and as Yaron alluded to, this was always less important to Ionis—is that in this study population, they tried to isolate Lp(a) as a variable. So other factors that could contribute, like blood pressure and cholesterol, needed to be under control.

But I think the flip side is that trying to tease out Lp(a)'s impact on outcomes in the context of patients with other well-controlled risk factors might also have its own intrinsic risks. Or, maybe said another way, maybe it isn't that big of a risk factor when other things are under control. How do you think about that? And are these other studies done the same way, or could they be different tests of this in a different context?

I mean, I think there are definitely different contexts. I think those are nuanced differences, though. Certainly, there's a chance that, given the better activity, different patient enrollment, and different care under the clinical trial, there could be differences in outcomes.

But again, what we're probably going to see is that you're going to have a very well-controlled patient population if you're not prospectively forcing that into the study. It's just going to happen because of clinical care. If you're going into a clinical trial, it's hard to imagine that you're not going to get the best care. So even if you're not in the best state going into the study, you're probably going to be managed to some meaningful extent. And that could potentially make it even harder to really see a placebo effect.

But again, they are different studies. They're different drugs. You can't fully write off the Lp(a) thesis. Look, you and I, we've been through this with beta-amyloid. For a decade, I was writing off beta-amyloid drugs, and now I'm a convert. The drugs that are getting brain penetration and getting plaque removal to a meaningful extent seem to actually do something, right?

So there are cases where the biology eventually works out in favor of pharmacology. I just don't see the differences here as so substantial that I think they're likely to play out favorably.

Maybe—

Yeah, makes sense.

Maybe a bigger question for cardiovascular development. Between the ZEUS trial failure, which everyone was kind of watching for—hsCRP and its link to outcomes, and how inflammation could impact cardiovascular disease—and now Lp(a), any general commentary on what this means for cardiovascular drug development?

I mean, to the last point, I think the challenge is that the bar is now a lot higher when you have to enroll patients that are very well controlled otherwise. The question is, can you really enroll sort of a random population that is not well controlled and run the study that way? Maybe it's going to be more real-world. I don't know. I don't know how—then how do you convince payers to give you a premium price, though?

Obviously, it's going to be a lot easier to control cholesterol with generics and, obviously, LDL and blood pressure. But I think the biggest challenge, though, is that the powering of the study is going to be based on the historical data with epidemiology, which is not really predictive given how well controlled they're going to be now. And maybe that's really the challenge: How do you write the right study?

What's interesting is that both Amgen and Lilly—and, look, they're both very legitimate players in these areas. They know cardiovascular disease really well, and yet they knew that these events were happening slower than expected. Yet Amgen, literally 6 months ago—actually, 8 months ago—launched a primary prevention study with Lp(a), obviously a huge study, and Lilly almost doubled their primary/secondary prevention Lp(a) study.

Novartis, as you know, also has what is expected to be an annual Lp(a) treatment that is moving to a pivotal study. We'll see if they cancel that now. But they all decided to continue forward despite knowing that the events are not looking great.

Yep. Thank you for that great commentary, Yaron. I think it's interesting, just flipping to what this means for some of the private companies. There's a number of companies in the private-company landscape that are working on Lp(a), potential combinations with Lp(a).

We've seen a number of dual-siRNA approaches. I think the hope was that cardiovascular disease could be an area where it would become more accessible for smaller biotechs if we could rely on biomarkers that lower the cost of capital in pretty early-phase trials, in order to raise the money needed for these CVOTs. That story just got a lot more complicated—not just with the Lp(a) failure, but also with the ZEUS failure, which calls into question the predictive power of hsCRP.

We're back to cardiovascular being really hard and taking some time to get the data that we need to understand...

The therapeutic impact. Let's continue with Novartis, actually. Brian, maybe I'll go over to you. On Friday afternoon, there was the Lp(a) update, and then we got the update on DM1. Why don't you talk about that with us, Brian?

4. Avidity's DM1 Bet Unravels

For me, Novartis is the gift that keeps on giving over Labor Day. I cover Dyne, so this was an important data set, as they have a competitive DM1 product with a lot of similarities. We could copy and paste a lot of what we were talking about with Lp(a) to DMPK knockdown, which is the target of these drugs to help correct splicing in patients with DM1. It's a large muscle disease and an orphan indication, but relatively large for an orphan muscle disease.

This was another big issue for Novartis. They did this $12 billion acquisition of Avidity less than a year ago. Avidity is a platform, and they have a couple of things going on, but they certainly highlighted this as a sort of crown jewel. I think everyone thought of this drug as the main driver of Avidity's valuation last year, when they did the deal. Novartis even put up a slide with maximum potential peak sales of about $6 billion, sort of justifying underwriting the acquisition last year.

This phase 3 failed. Novartis spun it as maybe a little more than just being a flat-out failure, unlike Lp(a), but it wound up not working. This has been a hot debate among the biotech community: whether or not this study was going to work, whether or not these drugs work, and how to properly design studies to hit endpoints that are relevant in DM1. I think it does call into question the idea that you can knock down DMPK in muscle.

This is a muscle-targeted silencing drug via an antisense oligonucleotide against DMPK. They have shown DMPK reductions in patients and decent benefit on an endpoint called vHOT in phase 2. That's what they designed this phase 3 around as the primary endpoint, with a number of secondary endpoints. Novartis teased that there might be something interesting in subgroup analyses or secondary endpoints in this study, which we'll have to see when they present the full data.

The failure is a huge disappointment for DM1 patients, who don't really have an option in terms of care or treatment. This would be disease-modifying if it worked, and it raises a lot of questions about what happens for DM1. I never got there on Avidity. I started covering Dyne and not Avidity because I thought there were a lot of questions about the next downstream marker in terms of what you're looking for as pharmacological activity.

There's this measurement, CASI-22, which is a measurement of splice correction. It's not totally standardized, and there is a lot of assay variability and all the questions that you always have when you're measuring something like splice correction in patients. The Avidity data never really showed me any compelling evidence that they were actually achieving splice correction, despite showing DMPK knockdown in blood measurements. Whereas Dyne does show placebo-adjusted CASI-22 benefits. Their average is about 25% splice correction.

That's where I was going into this data set. It certainly causes me to question the overall thesis: Do you need a lot of splice correction or a little splice correction? Is this the right assay? Is there so much variability that you don't really know what's going on in these patients? The bigger question is whether vHOT is just one of these endpoints that's so heterogeneous over a long or short period of time, or in different scenarios, that it's very hard to show a benefit on vHOT. Maybe this is just the wrong endpoint to use for DM1 studies.

Companies can think about looking at different endpoints over longer time frames to show functional benefit. We're going to see expansion-cohort data from Dyne early next year for their version of this drug. We're also probably going to see the full Novartis data set at WMS in Japan in a few weeks, so we'll get a better picture there.

Certainly, this is another big disappointment from the standpoint of Novartis management, given the scale of that acquisition. I know Yaron was going to talk a little bit about seeing some activism against Novartis in the wake of these events.

Do you guys mind if I just chime in on the DM1 space real quick?

Go for it.

Do it.

I cover Dyne too, like Brian, and I've covered Alnylam and Ionis for a really long time. As Brian alluded to, a big question with the Avidity approach in DM1 was that they use siRNA, whereas Dyne uses an antisense oligonucleotide. In most readouts where the two are head-to-head, siRNA has outperformed. We just saw this in the TTR space. It's probably not just a payload issue, but the siRNA payload seems to be more potent and may have a wider therapeutic index most of the time.

This disease is different because the protein isn't toxic; it's the DMPK RNA that's stuck in the nucleus of the cell. There was always a question as to whether siRNA is the right modality for that, because the original description of the siRNA mechanism of action was that it was cytoplasmic. What's interesting is that if you look at a couple of the diseases where ASOs have thrived, they've been nuclear targets, with SMA being one of them. We don't see a siRNA SMA program in clinical development.

When these companies were private, before they went public, and when they were early public, they had really only preclinical data. Even the Dyne preclinical data, we always thought, was much more convincing for nuclear knockdown. Some people I talked to on the industry side who were involved in siRNA were a little skeptical—not that siRNA wouldn't work at all, but that it was really the right modality. If it's active in the nucleus but much more active in the cytoplasm, it's going to lower more of that healthy DMPK transcript, which is probably half of it.

To Brian's point, we've seen DMPK lowering for these companies, but you can't differentiate between the mutant and the wild-type transcript. I'm still hopeful that Dyne has a shot. The endpoint question around vHOT is part of the right question, for sure. If you look back at the phase 1 and 2 data for both companies, Dyne's vHOT data does look more consistent from cohort to cohort. They have a bigger impact on splicing.

We saw Dyne's stock trade down a lot and then recover a decent amount, too, because I do think some people appreciate these nuances. For Dyne, if it does work, the setup has changed a lot. Previously, they were a second mover trying to pursue an accelerated-approval regulatory strategy, while Avidity was pursuing full approval. Now that could be different. If they actually do work, they're in the lead, and their upside, I think, has increased.

Yeah. Maybe let me kick it off, and then Brian can chime in and supplement further.

5. Novartis Faces Activist Pressure

This is fairly unusual. The key letter came out of Artisan Partners, which essentially raised concerns about board oversight of acquisitions, whether Novartis needs stronger board talent and an acquisition committee, and advocated for a compensation overhaul away from adjusted metrics that exclude specific write-downs, which are usually related to one-time acquisition-related charges. Ultimately, it called the party, sort of quote-unquote, “being over” and really criticized the chair.

It sounds like several other shareholders spoke with Reuters and also really called into question the acquisitions and the merit of some of them, whereas others defended him, saying this is within the normal sort of odds of development. I think Novartis basically said that its strategy will remain unchanged. They’re going to be looking at the broad pipeline, capital allocation with bolt-ons, dividends, and buybacks.

I think what was outsized here is that the deal for Avidity was about a $12 billion deal. Usually, Novartis talked about $5 billion and below for bolt-ons. I think the challenge, if you look at Street models, is that our team specifically on the pharma side—our esteemed colleagues—removed about $5 billion in sales from the model in their published research. It shaved off the EPS CAGR by about 200 basis points, to 4% long term.

Where this really hurts Novartis is that these cuts came at the time when they needed these revenues to be there, as they’re facing loss of exclusivity. Again, I think this calls into question—Novartis actually hired a very esteemed analyst, Ronny Gal, to run BD and strategy there. It’s always a question of how much risk you want to take. Do you want to go early, maybe pay $2 billion, $3 billion, or $4 billion for bolt-ons before, let’s say, proof-of-concept Phase 2 data comes out?

At that point, you should be hitting success rates like a three-point shooter. I think it’s maybe 35% or 40% these days with the modern NBA, as opposed to $12 billion, like in this case, in which case you need to be a free-throw shooter at a 90% or 92% range.

We’ve always advocated in our work that when you do a $12 billion deal, you might not take, quote-unquote, “clinical risk”—famous last words, because you always do—but you never actually hit your commercial published sales estimates in the 10-K filings that you put out to the SEC. We’ve done an analysis, and rarely do these deals actually get remunerated because, as you know, no one’s going to say, “We pay 10 times peak.” What they say is, “We pay 3 times peak,” which is digestible and acceptable, but then they totally inflate what these revenues are going to do.

Look, the bottom line—and again, our view—is that we all know pharma is not a sustainable business model. You do have exclusivity, which is unique; no other businesses really have that. M&A is usually value-defeating. We do believe you should go early. Do 6 deals, not 1. Maybe hope for 2 of them to work, and then they’re going to work in an outsized way, and you keep the financial upside.

Just doing stock buybacks and acquisitions has historically never worked for large-cap biopharma, in our view. Something else needs to go right for these stocks to then come back. But Brian, over to you.

Yeah. I’ve been talking about that with investors a lot this week, too. Novartis just isn’t the only one. We’re looking at, I don’t know if I want to say unprecedented, but borderline-unprecedented losses of exclusivity across large pharma over the next 5 or 10 years. There is a need to replace these revenues.

I mean, look, that’s probably why we were having one of the best M&A years ever in biotech, and I think that’s a driving factor. These company management teams realize they can’t cliff their revenue stream, and they need to find things to fill in there.

I guess the question for me, and what I’ve been talking about with investors, is that Novartis has seemed like one of the more risk-tolerant companies in terms of buying things that aren’t completely clinically validated or de-risked. Certainly, with the Avidity deal, that’s kind of blown up in their face.

We all know—I have gotten blown up on so many clinical-stage companies before when I think something is going to go right and it winds up going horribly wrong. That’s the game. But for Vas and Ronny, they’re kind of in the hot seat just from the one deal. There was a probability that this deal was going to go south, and they took that risk.

I just wonder if this pushes more companies to do de-risked acquisitions. Is this favorable for early commercial or post-Phase 3 companies, where maybe pharma is going to just overpay to acquire revenues? You’re probably not on the hot seat 9 months later because a Phase 3 clinical study from the $12 billion deal you did winds up blowing up and having to be written down.

Well, yeah. Brian, DM1 is muscle, but it’s neurology, right? It’s a disease with subjective endpoints. We saw what happened with Cerevel and AbbVie, and we’ll soon see what happens with Karuna in the ADEPT study with Bristol. If that’s negative, I would say in neurology, development-stage deals are going to be a pretty tough sell until something’s fully de-risked.

Yeah, agreed.

The other thing that we’ve seen is a fair amount of fairly early M&A, where the deals are small enough and the company is early enough that if it does blow up or doesn’t end up working, you haven’t lost too much. It’s interesting, but I wonder if that could lead to a kind of barbell approach, where you’ve got to be pretty de-risked or early enough and cheap enough that if something does blow up, it’s all right.

Hey, Tess, can I ask you a question about that?

Yeah, please.

You and colleagues of yours often have board seats in earlier-stage companies that are probably dual-tracking the whole IPO-versus-M&A process. What is the conversation like inside an early-stage company when there’s a bid? How level-headed are some of these entrepreneurs, who are obviously deep believers in what they’re doing, about what the true probability of success is?

An investor and analyst might be involved in an IPO of a company and think it’s a really attractive stock, while still modeling a 30% or 40% probability of success. Is it hard for some of these earlier-stage company entrepreneurs to really think that way? What’s your experience?

I think it varies so much by individual and also by board member. It’s obviously an area where a lot of people have strong opinions.

For a lot of entrepreneurs, when they get a bid and an offer, that is real money on the table. You don’t want to ignore or diminish that, and you have to take it seriously. Where we find ourselves working with a lot of our entrepreneurs who have had offers like that is helping to reinforce confidence, because they’re there for a reason, and we’re investors for a reason: We believe that these are programs that can have value and that we can take forward independently.

You just have to really balance that, have a realistic assessment of what a path forward looks like, and really help reinforce the confidence of the executives on the team about the ability of the investor syndicate to stick by them and help bring things forward. That’s where we often find ourselves in these conversations. Of course, in every case, there’s a number where it works, right?

Makes sense.

Let me maybe chime in on this, and everybody, please chime in more. One of the things is that we spent a few years in the CFO role and even had some BD responsibility, and the team around where we were was very experienced in BD, so we got to learn a little bit about this area.

One of the things that may not be totally appreciated externally is that when these companies do deals, you know, when you get married, you marry who you want to. When you do deals, you sometimes ultimately buy who you’re able to buy. We always think on Wall Street, “There are 37 companies you could buy. You should buy these 3.”

Those three might not be sellers, to Tess's point. They might not be sellers at that point, either, or at the valuation that you want. So, on the other side, it's very hard to do Vas's job and Ronny's job. You don't have an unlimited set of options.

We always think these companies have 100 levers they can pull. That's not the way it works at all. When you need to do deals that are accretive to you at a certain point and at a certain risk parameter, you really don't have a huge number of options that are also in your therapeutic area, that fit, and that you're going to get consensus on internally—from R&D, finance, strategy, clinical, regulatory, and your board—to actually do. We have to keep that in mind, too.

Yeah.

I'm not defending; I'm just—

Really important point, Yaron. Yaron, maybe sticking with you, let's talk about one of the positive data sets that came out this week. Roivant had Phase 2 data in PH-ILD. Tell us about their data.

6. Roivant Delivers Positive Data

Yeah. That's really happy news, and big congrats to Matt Gline, who is a very frequent guest and the CEO of Roivant, as everybody knows. He's really done an outstanding job and is a real talent. Roivant is extremely well managed.

Recall that this is a program that is inhaled once a day. It's a soluble guanylate cyclase mechanism that they in-licensed from Bayer. When they did the Bayer deal, I think many of us were scratching our heads and saying, "Hmm, it's kind of interesting that Bayer developed this inhaled program," because Bayer has an oral drug that is approved for pulmonary arterial hypertension, or PAH.

What we're talking about is PH-ILD, which is pulmonary hypertension–interstitial lung disease. There are 4 different categories. This is sort of the 2 extremes of a different market from PAH. So why is Bayer out-licensing it? Ultimately, it's because they're clearly not moving forward in this area broadly as part of their rebalancing and restrategizing.

There was interesting Phase 1b data that showed a very good pulmonary vascular resistance benefit in the lung—reducing the blood pressure and the resistance in the lung, which improves heart function—and this is sort of the proof-of-concept Phase 2. This was about a 16-week study with, obviously, an extension—a decent-sized study. We all knew that PVR was going to work because, in the past, they showed a 38% reduction with a single dose. So we sort of knew that was going to work. It showed 56%, by the way, so really solid.

But we didn't really know whether the six-minute walk test would work. That is what you need to show in Phase 3. When you looked at that, you sort of knew physiologically that if you do one, the other one should follow suit. But there isn't really that much data to draw on in this area, either from them or from their competitor, Tyvaso from United Therapeutics, which is approved.

Specifically with Tyvaso, they showed the six-minute walk data, but interestingly, they never released their PVR data in Phase 3, so we couldn't at least have some correlation. The study, of course, going back to Roivant, was pretty small and underpowered for the six-minute walk.

That data came out and looked amazingly good: 36 meters. Thirty meters is what you need to see, and, more importantly, it got better and better, reaching a 50-meter benefit by 48 weeks. So patients continued to improve. In true Roivant fashion, when they released the data, they said, "Aha, surprise, the Phase 3 has already started."

They are constantly able to move timelines up, which was the original premise of Roivant: that they would do things better and faster than pharma does. So it was surprisingly positive that the Phase 3 is already ongoing, too, in what could be a sizable market. Based on relatively small penetration, you get to about 2.5 billion in sales.

They're also running a study now in combination with Tyvaso, which is the United Therapeutics drug. The one issue with Tyvaso is that it does cause cough, whereas the drug from Roivant does not cause cough at all, so it seems to be better tolerated.

Phase 3 is ongoing. They'll talk to the FDA to see whether they need to change the trial design, maybe just incrementally now that they have the full data in-house. The stock has performed really well. They're obviously launching brepocitinib now, which just got approved for dermatomyositis, and the expectation is for a solid launch. They have a lot of catalysts coming. So again, congrats to Matt and the team. Really great job.

Yeah. Just echoing how it's been a great year—a really strong year for Roivant, really strong execution—and it's fantastic that they not only released positive results but are also getting going on their trial really quickly.

I think the expectations for this program were very low, right, Yaron? I think it was a pretty big surprise that it had the therapeutic impact that it did. So let's move on to the next one. Paul, I think there are a couple for you. Maybe you can share some updates for Biohaven and Pharvaris that came out this week.

Yeah, sure. I can zip through these. Biohaven had what looks like a small setback but might be more meaningful in context for their KV7 drug in epilepsy, which was put on partial clinical hold. It's in Phase 3, and the data are coming quite soon.

There's some opacity around the disclosure here. It's related to a metabolite in rats where there's insufficient information to characterize the risk in people. We've seen things like this before. These can be significant issues, or they can be complete non-issues. There was a whole metabolite question with lumateperone back in the day in one species, and look at what that ended up meaning to Intra-Cellular Therapies: a whole lot of nothing.

The broader context here is that these guys are second behind Xenon in the KV7 space in epilepsy. Xenon's clinical data and profile for azetukalner are pretty incredible. Their data are awesome. Efficacy is differentiated, and safety and tolerability are really good at their lowest dose, which still has respectable seizure reduction. So it probably just makes it tougher for them to compete in this market. I don't know if Brian had anything to add on that one.

Yeah, just real quickly. I think the interesting nuance here is that they just did this deal with SK Biopharmaceuticals a couple of weeks ago, right?

Right. Yeah.

This may wind up being nothing, but I've been fielding questions on Biohaven like, "Is SK going to pull out of this deal? Do they have a basis to pull out of this deal?" I think SK held a call yesterday, and they said they're talking with Biohaven about not closing the deal until after this is resolved. Biohaven is saying they could resolve this in a month. Maybe a month from now, it's fine.

I kind of agree with you: I'm a big azetukalner fan. I like Biohaven as a company for reasons other than BHV-7000. Partnering with SK, which already has a very good epilepsy drug launch under its belt, seemed like an ideal partnership. The fact that they're bringing in $400 million upfront from the deal, I think that's a huge win for Biohaven.

I do not want to see any risk to this partnership. I think that's where people are getting edgy: "You definitely don't want SK to wind up walking away here because it's unclear that, outside of some Goldilocks position, Biohaven with BHV-7000 is going to be in a position with the dose flexibility to really compete with azetukalner, given how much more data they have."

Yeah, that's great context. Thanks, man.

And then, Pharvaris—Yaron alluded to it earlier—but these guys finally broke through something that people have been trying to do for a long time now: generate really competitive efficacy data with an oral drug for HAE prophylaxis.

HAE, or hereditary angioedema, is a disease where you have intermittent swelling attacks. As far as rare diseases go, it's probably become one of the better-served markets, just with the amount of iterative innovation there. Going back to Dyax and ViroPharma, and then today, you've got prophylaxis drugs like Takhzyro from Takeda, which is reducing the frequency of these attacks by 90%.

We saw BioCryst come out with the first oral prophylaxis drug, but the attack reduction there was in the 45% range. This was over 80%. I'm sure the analysts covering Pharvaris closely are trying to do the granular patient modeling or reconcile where this drug fits in an increasingly competitive landscape.

And there are other programs coming here too that could potentially further decrease the injection frequency, even to every 6 months. I think that will be an interesting question: Do patients prefer an oral every day, or do they prefer an injection every 6 months?

But as it relates to Pharvaris, the thesis looks like it played out. That company had a bump in the road, too. I think at one point they had at least a partial clinical hold. Look at the data now. Big success for them.

7. Structure Tests Oral Amylin

Yeah, great. Maybe we'll close with another data update that we got this week, which was from Structure Therapeutics. A lot of people were watching their oral amylin for weight loss. There's been a lot of excitement about the amylin space, particularly amylin injectables, and Lilly's eloralintide is probably the only amylin out there that could get to competitive monotherapy efficacy.

There's quite a bit of interest in Structure Therapeutics for having the first oral amylin that we now have clinical data on. Structure shared the results of their single-ascending-dose study, and the top dose, which was 10 milligrams, showed 3.3% body-weight loss with just 1 dose. I think the market was a little disappointed. The stock sold off, rebounded a little bit, and then sold off a little bit more.

Overall, some of the reasons for that may just have been higher expectations for overall weight loss or perhaps tolerability, which had pretty high rates of nausea and vomiting at some of the higher milligram doses. But I think it's exciting to see more oral drugs come into the space and see not just the GLP-1 orals. We have oral Wegovy, which was launched a while ago by Novo, as well as orforglipron.

Structure is right up there with their GLP-1 and now has another target to think about in terms of combinability with their oral. Any other comments or data that folks want to comment on from this week?

Great. Let's hope for—I think we had a good discussion. Lots of challenging data this week. Good to have a couple of bright spots, with the Roivant data in particular. I really appreciate everyone tuning in, and I wish everyone a great weekend.