Sean Kenney: Hi, I'm Sean Kenney, and welcome to the MFS All Angles podcast. Each episode, I'll sit down with portfolio managers and analysts to dive deep into industries, explore key themes for investors, and uncover how they approach investing from all angles. Our goal is to make the complex world of investing a bit simpler, and provide you with practical insights you can put into action.
Today I'm joined by Matt Scholder. Matt is Co-Chief Investment Officer of Equity. He's been with MFS for over 15 years, and for the past 12 years, Matt's been a research analyst focused in the pharmaceutical industry. Matt also is an avid music fan, he's logged over 70,000 minutes on his music account. So Matt, maybe we'll dig into that at some point, but I want to-
Matt Scholder: Sounds great.
Sean Kenney: ... welcome to the podcast, and thanks for spending the time with us.
Matt Scholder: Thanks for having me.
Sean Kenney: So our topic today is AI, artificial intelligence in drug discovery. And when I think about complex industries, one of the first industries that comes to mind is pharmaceuticals. So I'd be really curious from your perspective, you spent over a decade in that space. Just as a starting point, what excites you about the pharmaceutical industry today?
Matt Scholder: I think what excites me the most is, you're right, it's complicated, there's a lot of big words, it's science driven. But actually at its core, being a pharma investor is quite simple.
So when I first started 12 years ago, I had this epiphany and I thought to myself, drugs drive drug stocks. All I need to do is learn about the drugs and I'll get the drug stocks right. Fast-forward about 10 years, what I realized is, it's actually not drugs that drive drug stocks, it's new drugs that drive drug stocks. And so in my 12 years of doing this, I have sort of fallen in love with the process of analyzing pipelines, analyzing new drugs to try to handicap their commercial success, to ultimately pick the best drug stocks for our clients.
Sean Kenney: Yeah, interesting. And if you think about your 15-year history of investing, give us a little bit about your story, your investment story. And how did you end up in the pharma space?
Matt Scholder: So, I ended up in the pharma space pretty reluctantly. So I started MFS in 2010, I covered regional banks, and then at some point-
Sean Kenney: [inaudible 00:01:57].
Matt Scholder: Ted Maloney, our current CEO, he was then Director of Research, came to my office and said, "Matt, I'd like you to cover biotech and pharma." And I said, "No, thank you," because I had thought to be a biotech and pharma analyst, you had to be a doctor. And I assumed that the person who was doing it before was a doctor with the way he just threw out molecule names and the ease with which he interviewed management teams. And then Ted said, "No, actually, we think being a financial analyst first is the most important thing and we think some firms that prioritize the science and make bets purely based on that don't have the best track record. And so, we want you to be a financial analyst first. And no one's asking you to develop a drug, we're asking you to be a stock picker." And once I sort of relaxed into that thought and leaned into it, I ended up loving the coverage and it's what I've done for the vast majority of my career.
Sean Kenney: Yeah, that's so interesting because I think as a analyst first, I'm assuming you start by doing industry assessments and understanding the industry intimately. And one of the big parts of the pharma industry and any drug company is drug development.
Matt Scholder: Exactly.
Sean Kenney: Can you walk us through maybe really quickly, stage one, stage two, stage three, and how we think about that?
Matt Scholder: Absolutely. What drug companies are trying to do through that process is de-risk a molecule or a compound from a safety and an effectiveness perspective. In smaller patient populations in phase one, slightly larger patient populations in phase two, and then largest patient populations in phase three. By the time they get to phase three, they know the dose, it's been really de-risked from a safety perspective, so regulators feel comfortable giving it to hundreds if not thousands of people. And we're really trying to figure out, does this drug work? Is it effective? Does it slow the growth of a tumor? Does it lower someone's bad cholesterol? Whatever the drug is trying to do in the body, we're going to answer that question in phase three.
Sean Kenney: And so, why is it so expensive? I mean, if this is the most important part of the lifeblood of a company, it's also the biggest part of the expense structure for these companies. Why is it so expensive?
Matt Scholder: Huge amount of people expenses. First of all, PhDs are expensive, so you need to pay a lot of scientists to design these molecules in the first place. Running clinical trials is expensive, you need to pay doctors and nurses to go out and find the patients, to administer the drugs, you have to pay for the drug. In some cases, you might be running a phase three trial versus another company's drug. So you have to go out and purchase maybe a cancer drug that's $150,000 a year for 3,000 patients. It gets very expensive, very quickly. So drug companies are always trying to figure out optimal ways to increase the probability of success, lower the cost, fail quickly, and ultimately deliver innovations for patients.
Sean Kenney: And so your job as an analyst for MFS is to differentiate between pharmaceutical companies and drug companies that are likely to do well, versus ones that are likely to struggle. What are the investment implications of these trials and the expense and the R&D?
Matt Scholder: So what's fascinating about the industry is, all drug companies employ teams of PhDs. They're all incredibly smart, they all went to great schools. As an outsider, they all look about the same to me. Yet if you look at some drug companies, they could go through 10 or 15 years despite spending billions of dollars a year on R&D, and develop nothing. And then you can have a similarly sized company with equally smart PhDs and they can develop a drug that changes the world.
And so my job is not to pick the drug in phase one, that's almost impossible. My job is to look at the data once it hits phase three, once it's de-risked, before it starts generating revenue, and ask doctors and patients and do my own research. How big is this drug going to be? Is it going to change the revenue growth profile of this company? Therefore, it's going to drive outsized revenue and earnings growth and be a good stock.
Sean Kenney: And so when you think about that as an investment decision that you're making, and you look at AI and you say the future of this industry leveraging AI, where do you see AI fitting in from an application perspective for pharma companies and drug companies?
Matt Scholder: Absolutely. So let's go back to where we started, where the epiphany that I've had along the way, is unlike other industries where you might focus on margins or capital return or customers or competition, with drug companies, it's drugs drive drug stocks and new drugs drive drug stocks.
So with that framework in mind, what's so interesting about AI is I wanted to ask the question, how could AI change the probability of success for either more new drugs to come through the pipeline or better drugs to come through the pipeline? And one of the best parts about working at MFS is we have access to management teams from every pharmaceutical company around the world. And so I've been able to ask the heads of R&D from almost every drug company this question, "How are you deploying AI in your labs? How are your researchers using it today? How might it change in the future? What are the weaknesses? What are the ways in which can be improved? And ultimately, what does this mean for your business?"
So even just earlier this week on Tuesday, we had the head of R&D of a very large pharmaceutical company, where we spent 10 minutes discussing this exact question, and I can share some insights if you're interested.
Sean Kenney: Yeah, that'd be great. I mean, I'd be curious in particular, maybe we start there. Are these companies putting their own talent in the building or are they using third party technology companies to do this work?
Matt Scholder: So the first thing that's become clear is the advantage that pharmaceutical companies have is all of the data based on all of the drugs they have in their pipeline today or in their base business today and the years of running clinical trials with those drugs. So you can think of pharmaceutical companies as data rich, and the tech companies are rich in AI talent. And so they've tried to figure out ways to partner with tech companies so that they can harvest the AI talent that's outside the pharmaceutical company, because they're desperate for the data that the drug companies have. And so that balance of how every drug company's handling that varies depending on the company. But what's become very clear to me in talking with all these management teams is that those who have the richest, deepest data set will be able to extract the most value using AI going forward.
Sean Kenney: And so do you see this in the short term being a cost to the companies? I mean, I'm assuming they have to invest in AI, and so does this become a short-term cost that ultimately they realize long-term benefit?
Matt Scholder: The most interesting thing is when you talk to other management teams in other industries about AI, they tend to talk about saving money first. It's going to make us need fewer employees or our back office operations more efficient. When you talk to drug company executives and heads of R&D, they acknowledge there are absolutely some cost savings. For example, a regulatory submission that might take 10 people in a month to put together, you can now do with one person in 15 minutes, but that's sort of a little exciting.
What gets them really excited is developing drugs that were previously un-druggable targets. And so what that means is, a scientist may know that something in the body causes cancer or causes some disease to progress, but they've been unable to develop a drug that targets that molecule which causes the disease progression. Using AI, their chemists can design molecules more quickly and potentially more effectively to target what have been previously been un-druggable targets.
Sean Kenney: Yeah.
Matt Scholder: And that's what gets them excited.
Sean Kenney: And so if you think about the winners and losers in that space, are there any themes that you're seeing in terms of how some companies are really either investing or understanding the utilization of AI versus others? Is it a, being bigger is better situation?
Matt Scholder: So it's still early days, but the biggest insight I've had so far is this idea of potentially a virtuous cycle, where a company that really specializes in one therapeutic area that has the world's largest drug in that therapeutic area and has therefore the most patient data and years of clinical data, will be able to use that to potentially create new versions of that drug, combinations with that drug. Or you think about deploying that drug in new ways to treat other illnesses, and that would be very unique to that one company. It's going to make it harder for startups, it's sort of an incumbency advantage, and it's those with deep therapeutic expertise that will be able to harness those benefits.
Sean Kenney: And from an industry perspective, let's talk about the size. How many new drugs are coming to market, and what's the probability of success for a new drug coming to market today, and how might AI change that?
Matt Scholder: So that's the other angle, which is a little bit TBD, but is really interesting to think about. So every year, between drug companies and universities, about 2,000 drugs start the clinical trial process. By that I mean they go to the FDA, they submit an application and they say, "We would like to study this drug in patients." So think of 2,000 as the beginning of the funnel. And then every year, the FDA is approving about 40 or 50 drugs. So it's a very low single digit probability of success going from 2,000 to 40 or 50.
So the other question we're asking is, is that 2,000 number with the help of AI and the speed of drug development, going to look more like three or 4,000 in a few years, and therefore on the other side of the funnel it could be 50 or 100? And if it does, what does that mean for payers? What does that mean for healthcare budgets? What does that mean for contract research organizations? What does that mean for tools companies which provide services and facilitate R&D through their business? So we're thinking about all the downstream impacts of how that funnel could change.
Sean Kenney: And so, one more downstream impact is consumers.
Matt Scholder: Yes.
Sean Kenney: What does that mean to consumers? Is this lower drug pricing or is this more innovative drugs?
Matt Scholder: So lower drug pricing is a hot topic, as it is in every administration. I would think for consumers, I would be more rooting for innovation. So if I were a consumer, I would be more excited that cancer in 10 or 15 years might be a disease that you live with rather than die from, as opposed to the cost of my medicine is going to be significantly lower. I think drug pricing will be resolved through legislation and government action. We should really think of AI in drug development as driving innovation and better outcomes for patients.
Sean Kenney: Yeah, I mean, that's a narrative that's important to differentiate, right? Because it isn't necessarily just about cost-cutting, it is about the patient outcomes and innovation.
Matt Scholder: It's one of the best examples of using AI to actually make the world better and healthier, as opposed to just more efficient and saving money.
Sean Kenney: Yeah. One thing that you touched on that was interesting to me is the downstream implications in the ecosystem, whether that's within the healthcare ecosystem more broadly, or even in the technology ecosystem. I would imagine as an investor on the MFS investment platform where we work together in teams across sectors and industries and regions, you get some unique insights into the second and third derivative impact of all these innovations. Anything come to mind that jumps out?
Matt Scholder: I think frankly, it's something that I think we're really well positioned to do. So we've had a number of sector meetings that have explored this topic from other angles. And the way we've approached it is we have C.V. Rao on our team, who's a PhD and an expert in all the technological and complex nuances of AI in drug development. He walked us through everything he knows. And then all the analysts on the team from the person who covers managed care, the person who covers tools and life sciences, to me as the drug analyst, we have then thought through, what are the implications for our companies if clinical trials are larger, if there's more clinical trials, if they're cheaper, if there's more drugs going through the funnel? What does that mean for more drugs being approved? And we've tried to understand, what are the watch points, what are the guideposts we're going to be looking for over the next three to five years to think about how this new technology could change ultimately the revenue and earnings for the companies that we invest in?
Sean Kenney: Yeah, does the technology team have skin in the game here? How are they thinking about this, leaning into this with you?
Matt Scholder: So as healthcare investors, we know a lot about drug development and very little about the technology, so that's another advantage we have. I'd highlight two things. Number one, all the analysts here, most of the senior analysts are on two sector teams. I happen to be also on the consumer team, and I'm involved in some very large technology companies that are at the intersection of technology and consumer. So given my own expertise and background, it's been helpful to lean on the other parts of my coverage in this analysis. And then of course, we're fully integrated as a team and our C.V., who I just mentioned, sits on both the technology team and the healthcare team and all the sector teams. And we'll lean on our software analysts and our hyperscaler analysts to understand how large language models are developing, how the costs are changing. And we all sit on the same floor. In fact, the hyperscaler analyst is just two doors down for me, so it's easy to just knock on her door and ask her all my dumb questions about AI.
Sean Kenney: Yeah. Well, I'm sure they're not dumb, but I'm sure she gets answers.
Matt Scholder: They're pretty dumb.
Sean Kenney: Yeah.
Matt Scholder: Relative to what she knows.
Sean Kenney: That's right.
Matt Scholder: And then she sometimes has questions about drug development, and that's one of the great things. In fact, just a quick anecdote, one of the companies she follows, one of the new management team members is from a drug company. And so, one of the first questions she asked me is, "Did you interact with this executive when they were there? What did you think of her?" And that's the type of conversations we're having all the time because we know each other, we spend time together, we're in sector meetings together, we sit near each other. And I think that's MFS integrated investment platform at its best.
Sean Kenney: How important is it to talk to leadership? I mean, is this stuff you can't just read in an annual report or the science report?
Matt Scholder: The management teams?
Sean Kenney: Yeah.
Matt Scholder: I think it's incredibly important. I think it's one of our differentiators. Every management team that comes to Boston to meet with investors, with very few exceptions, spends time with us. So, there's management teams in the office right now as we speak. And then we make a big effort to go out and see companies at their headquarters and visit them on site and see multiple layers of management, not just the CEO or CFO, but the people actually running the business, running business units to try to understand them better.
So just to give you an example, I think 18 months ago, the healthcare team, we did a trip all around Boston to visit all the life sciences tools companies at their headquarters, tour their labs, meet their management teams, understand their science, dig into the technology. And it was a group of 20 or 25 of us for a day and a half doing nothing but that. And there's a real difference between visiting companies on their home turf, where they're a little bit more relaxed, you can learn about their business in a more nuanced way. Frankly, they're grateful that we take the time and the interest to go see them, and they're a little bit less scripted, they're not just going from one investment meeting to another and we can really dig into the strategy and the business and get some unique insights, which we then turn around into the position sizes that we take.
Sean Kenney: Yeah, yeah. So, last question. If you think about your job as an analyst is to differentiate between winners and losers, we touched on that a little bit. If you were to think about this industry, does AI raise all boats, or will there really be true winners and losers or firms that take an outsized share of the benefit of AI?
Matt Scholder: So I would acknowledge up front, this is still TBD. This is cutting edge technology, we don't know how it's going to all play out. But my initial thought is this is going to make the bigger company stronger and the winners keep on winning. Because if the most important thing for any large language model or to extract value from AI as data, it's the big companies that have that data that are going to be able to deploy it. So I think there's a real incumbency advantage that could develop and grow over time, which is something we're going to be thinking about as we're investing in this space.
Sean Kenney: Yeah. Well, just to summarize, just for a layman investor, someone who doesn't sit in the pharma space every day, what you said was the real benefit of AI, yes, there's likely some cost efficiency, but it's really the innovation and the improved patient outcomes. That's the potential and the excitement about AI, is that right?
Matt Scholder: It's what excites me as an investor, because again, coming back to where we started, what's the most important thing for a drug analyst to focus on? New drugs. What could change that? AI in drug development. So we're spending a lot of time trying to understand it, trying to dig into it, because if we can understand those trends and increase our probability of identifying those big new drugs as they come to market, we're going to hopefully find stocks that outperform for our clients.
Sean Kenney: Yeah, and then to take that one step further, it's, AI can mean more innovation. Innovation means great new drugs, and great new drugs mean better stock prices. Is that the simple formula that you'd follow?
Matt Scholder: It's hard to think of a drug stock that has outperformed without a great new drug driving it.
Sean Kenney: And maybe the final point is, it's not going to be equal. There will be winners and losers. There will be firms that take an outsized benefit from this versus others.
Matt Scholder: And our job is to roll up our sleeves, understand that, take risk, and dig into this in as much detail as we can to try to identify those winners and own them over very long periods of time and hopefully drive tremendous amounts of alpha for our clients.
Sean Kenney: Yeah. Well, we really appreciate the time, Matt. I do want to finish with one final most important question, which is, 70,000 minutes.
Matt Scholder: Yes. It's a lot of music.
Sean Kenney: On your music account. I did the math, that's over three hours a day, so hopefully you're sleeping. But what would be your go-to today?
Matt Scholder: Okay, so what's interesting is, it really depends on what I'm doing. There's nothing better than working on an Excel spreadsheet and listening to deep house music.
Sean Kenney: Oh, so you're working and playing?
Matt Scholder: That might surprise you. And then when I'm working out, it's '90s hip-hop, which is what I grew up with.
Sean Kenney: Excellent.
Matt Scholder: And so I still listen to the same music I did in high school, even though I'm 43 years old.
Sean Kenney: Great. Well, we really appreciate the time. Thank you for the insights and thanks for being with us.
Matt Scholder: Thanks for having me.
Sean Kenney: Thank you for listening to All Angles. Don't forget to subscribe so you don't miss any episodes in the future. And until next time, consider investing with All Angles.
The views expressed are those of the speaker and are subject to change at any time. These views are for informational purposes only and should not be relied on as a recommendation to purchase any security or as an offer of securities or investment advice. No forecast can be guaranteed. Past performance is no guarantee of future results.
Unless otherwise indicated, logos and product and service names are trademarks of MFS® and its affiliates and may be registered in certain countries.
65116.3