Evidence-Based OncologyTM sat down with Andrew W. Lo, PhD, to discuss his proposals to increase investment in medical innovation and utilization by giving investors more ways to support both research organizations and individual patients.
Andrew W. Lo, PhD, the Charles E. and Susan T. Harris Professor at the MIT Sloan School of Management, has a wide variety of research interests. He's best known in the healthcare world for ideas that could greatly increase the development of new drugs and access to expensive cures. The director of the MIT Laboratory for Financial Engineering, Lo is a principal investigator at the MIT Computer Science and Artificial Intelligence Laboratory and an affiliated faculty member of the MIT Department of Electrical Engineering and Computer Science. He is a past winner of the Battermarch, Guggenheim, and Sloan fellowships and has been named one of the 100 most influential people in the world by TIME.1
His proposals seek to increase investment in medical innovation and utilization by giving investors many more ways to support both research organizations and individual patients. For example, Lo sees opportunities for attracting new investment via securitization, the practice of combining individually risky assets into large pools that provide more predictable returns. The resulting cash flows are then divided into tiers based on levels of risk and reward favored by different investors.
Lo’s proposals typically rely solely on investors, drug developers, and patients acting in their own self-interest to produce the desired outcomes. However, he does see some opportunities for relatively small investments by philanthropies or the public sector to produce outsized returns in areas like pediatric cancer, where traditional finance tools would not be enough to attract socially optimal levels of investment. Indeed, by simply protecting private investors against loss during the riskiest phases of pediatric drug development, charities or governments may secure far more investment than they could fund on their own.2 This concept was presented in June at the American Society of Clinical Oncology annual meeting,3 with full results published in September in JAMA Oncology.2
Evidence-Based OncologyTM (EBO) spoke with Lo about his ideas, including those in his recent paper.
EBO: How did you get interested in how we finance drug development and usage?LO: It was really for personal reasons. A few years ago, a number of friends and a family member were dealing with different types of cancer, and it was through the process of trying to understand what they were dealing with that I realized that finance actually plays a pretty significant role in drug development.
For example, even though we seem to be on the verge of a number of breakthroughs in how we deal with these diseases, funding for early-stage drug discovery is actually getting scarcer. Why do we see this so-called “Valley of Death” for preclinical [research and development] and phase 1 clinical trials? In trying to make sense of this conundrum, I began doing research on the economics of the biopharma industry and realized that we could actually make a difference for patients if we used better methods for financing drug development.
EBO: How does that work?
LO: My conjecture, which now has substantial supporting evidence, is risk. Drug development is really at its riskiest between the preclinical stage and phase 2 clinical trials. As financial pressures have increased on drug companies and venture capitalists [VCs] to improve performance, the response has been to focus on better bets, bets that are more of a sure thing. This means waiting until drug development projects reach certain milestones before investing.
Another risk that drug companies face is the risk that—thanks to all the recent biomedical innovation that’s been going on—a better drug gets developed, destroying the valuable franchise that these companies have invested billions in. We see this happening now with gene therapies that are on the verge of curing certain diseases that used to be chronic manageable conditions. It’s going to be a very interesting market dynamic as we see patients choosing between one kind of therapy and another and what kinds of pricing policies will emerge.
Despite those risks, overall returns on drug development are high enough to suggest we’re not investing nearly enough in it.
EBO: Why is that?
LO: Drug development has 3 unique characteristics that, taken together, make it very challenging for investors. First, it takes a long time: typically 3 to 5 years before you hit the first major milestone and 10 to 15 years until you get an approved therapy. Second, it takes a large amount of capital to bring a single product to market, typically 1 or 2 orders of magnitude more than you need for start-ups in other industries. Third, the success rate is very low, about 5% in oncology for example, based on historical data.
There’s a small group of traditional biotech investors that are very sophisticated and understand how to manage these risks, but the larger pools of capital from ordinary investors that have been rushing into other areas like technology, social media, or cryptocurrencies simply aren’t there for this market. This is where the opportunity lies for better financing. If you structure the investments differently, you can make biotech attractive to more people and draw more money into drug development.
EBO: How?LO: There are actually 2 ideas that have been used for decades in other industries and demonstrated the ability to reduce risk and improve average returns for investors. The first idea is “multiple shots on goal,” to use a hockey or soccer analogy. If you put together series of investments into a single financial vehicle, you are, in most cases, going to be able to reduce the risk and increase the likelihood that you have at least 1 or 2 successes. And in biomedicine, you only need 1 or 2 approved drugs to pay for all the other tries and still earn significant profits for investors.
But to create a large enough portfolio of multiple shots on goal, you need [a] much larger scale than the typical VC would have— on the order of billions of dollars in the case of cancer therapeutics, rather than a few hundred million dollars. This is counter to the traditional VC view that “small is beautiful.”
Which brings us to the second idea: using different kinds of financial instruments to fund these multiple shots on goal. Traditionally, biotech VCs use convertible preferred debt and then eventually equity to finance these start-ups. That’s certainly the tried-and-true approach. But to create a large enough portfolio of multiple shots on goal for drug development, you need to attract more capital, and the way to do that is to offer different kinds of securities. You can divide the underlying portfolio of assets—claims on future drug sales of multiple drug targets—in ways that let you sell low-risk, low-return bonds to investors who want safety, higher-risk equity to investors who want to gamble for big returns and every other type of security for investors who fall somewhere in between.
EBO: You also believe that finance can increase utilization of expensive short-term treatments that either cure diseases or provide lifelong benefits. The idea is to find pools of capital that would be used to lend patients money that could be paid back over time.LO: Yes, but that’s not the new idea. It’s being done right now for all sorts of elective surgeries. Dental reconstruction is a good example. The typical cost of full-mouth reconstruction can range from $20,000 to $50,000, and candidates for these kinds of procedures can get consumer loans to pay for them right now.
What’s new in our proposal is to do this on a much larger scale, greatly increasing patient access to costly therapies, and to be able to securitize and package these “healthcare loans” and sell them to investors who are perfectly happy to take on large pools of diversified kinds of risks. Again, the investors get multiple shots on goal, but in this case, you’re talking about multiple consumer loans or loans to health insurers.
The idea really came about because of the possibility of curative gene therapies as well as other shorter-duration treatments like Sovaldi and Harvoni, 12 weeks of pills that can cure hepatitis C. Taking traditional medications is like renting an apartment: You pay for your benefit a month at a time and keep on paying as long as you want to live in that apartment. New medications like gene therapies are more akin to buying a house. You pay the whole cost up front and the benefit lasts for many years. But most people can’t afford to pay for the entire house in cash, so they get a mortgage, and that’s what we’re talking about here: drug mortgages. It’s the difference between “renting health,” one pill at a time, versus “buying a cure” and paying for it in installments.
The reason that this is an interesting approach is not just because it makes expensive one-time therapies more affordable for patients and payers. It can also serve as a way of dealing with cures that don’t really cure. We don’t yet know how permanent the effects of gene therapies really are; they’re supposed to be permanent, but we don’t have any real experience to rely on. What if it turns out that, after 2 or 3 years, the “cure” stops working and the patient relapses? Well, if you’ve paid for the therapy through a drug mortgage, you can simply stop making payments. That aligns the interest of the drug companies with the patients and the payers.
Based upon some very simple simulations that we’ve run by paralleling this submarket with the student loan market, we think there are tremendous amounts of resources that could be devoted to increasing patient access to these kinds of therapies. And with the 300-plus gene therapies currently in clinical trials, this is only the beginning of a huge wave of cures coming to patients over the next few years.
EBO: You have also explored the idea that insurance companies, rather than patients, would take out these loans.LO: Yes, the idea of stretching out payments over a period of time is pretty straightforward, but the big question is who’s going to be paying that mortgage, and the natural response is insurers. That’s why we have health insurance.
The problem is that insurers expose themselves to potentially huge losses when they make large up-front payments for treatments that provide their policyholders with decades of benefits. This wouldn’t be true if policyholders always stayed with the same company. If you pay for a cure for one of your policyholders and that policyholder now lives for the next 30 years instead of dying in 1 or 2 years, you’ve got 28 more years of premiums that you can use to offset the cost of this cure.
But in practice, that policyholder can leave her health plan at any time. Suppose she moves to another state after only 5 years. So now, instead of having 28 years of premiums, you’ve only collected 5 years’ worth; the remaining 23 years of premiums go to an insurer in another state. That insurer will benefit from a healthy policyholder who’s now cured of this disease, thanks to the previous insurer. She may have other problems, but that’s one really serious disease that the new insurer doesn’t have to pay for.
There are a number of ideas for solving this patient-migration problem through regulation or legislation to protect insurers. This is critical because insurers are going to be very reluctant to finance expensive cures until they’re protected against this scenario. And currently, they’re the only ones who have the resources to finance the most expensive cures.
EBO: Because the idea of lending money directly to patients doesn’t work when the loans get much bigger than $40,000?
LO: Exactly. There are very few patients that can afford million-dollar therapies. And we’re now on the verge of developing a gene therapy for hemophilia. The best guess today for that price tag is about $1.5 million. There are few enough people who can afford $1.5-million homes, never mind $1.5-million drugs. And in the case of a home, at least there’s some pretty substantial collateral. In the case of your health, it’s very difficult to repossess a lung, so that’s why we used student loans as a model for this market: You purchased a college education, but they can’t really repossess your math courses. That’s why, ultimately, insurance companies are the ones that will be taking out these loans, and we need to deal with the patient-migration issue that they face.
EBO: Talk about the research on how you could use these financial tools on pediatric disease, particularly cancer. Why doesn’t the market work now, and how can finance improve it?LO: Ironically, I think part of it is because we care so much about our children. Traditionally, pharma companies have not developed drugs for children exclusively. They start with adults for the simple reason that drug testing is a very dangerous business that often involves toxic side effects. Patients do die in clinical trials, so it’s a matter of ethical consideration to have adults take on these risks first before we expose children to them. But the consequence of that kind of ethical perspective is that pediatric oncology drugs are harder to come by. Drug developers will focus first on adults and unless, and until, they develop a successful therapy, they won’t try it on children. Now that’s changing, particularly with recent legislation requiring drug companies to develop therapies for adults and children simultaneously. But historically it’s been a challenge.
It’s also harder to develop pediatric oncology drugs because children have far fewer mutations than adults, implying fewer genetic biomarkers and druggable targets. Also, the biology of a 5-year-old can be quite different from that of a 12-year-old, and both are different from the biology of an adult, so it can be more complex to develop therapies for pediatric indications. Finally, the smaller patient populations make the economics of pediatric cancer drugs less attractive than those of adult counterparts, even with incentives like priority review vouchers and orphan drug designation. We ran the numbers and were shocked to discover that the simulated rates of return for pediatric oncology drug-development programs typically yielded double-digit negative returns on capital. Companies still do develop pediatric cancer drugs, but it’s probably more for ethical than financial reasons. And while it’s laudable that companies and investors still do invest in pediatric cancer drug development, they don’t invest nearly enough because the financial risk/reward profile isn’t very compelling.
Our proposal to get more investment into this field is a public—private partnership where the government and/or philanthropic organizations get involved in funding some of the riskiest parts of the development chain, preclinical R&D and phase 1 trials, and then hand it off to the private sector for completion after phase 1. We found that relatively reasonable amounts of public-sector funding, certainly much less than what goes into other areas of the drug development, could actually go a long way toward making the private sector interested in these kinds of health issues. Now that our paper is published (including our simulation software), our hope is that it’ll attract the attention of industry participants as well as philanthropic organizations and government agencies and that might spur them to collaborate with private-sector investors and biopharma companies.
EBO: If you were the Trump administration’s health policy czar, what’s the first thing you would do?LO: The first thing that I would do is to use the convening power of the White House to bring together various stakeholders to work together collaboratively to develop solutions to some of the biggest medical challenges facing our country. The 3 areas I would prioritize are Alzheimer’s, infectious diseases, and antibiotics. These 3 areas present huge challenges, both current and pending, not only for our country but for the world, and they’ve gotten far too little resources and attention to date. As we saw with the Biden Cancer Moonshot, the White House can be extremely effective in coordinating the efforts of various groups to deal with the biggest issues facing humankind.
EBO: How could things improve on the regulatory side?LO: Having now spent some time interacting with the FDA, I have to say that I’m incredibly impressed by the quality of their staff, their dedication, and how effective they’ve been. One of the most surprising things about them is their openness to new ideas and constant focus on improving “regulatory science.” To that end, I’ve been collaborating with several of their researchers to incorporate patient preferences into drug approval decisions. The standard approach for weighing the evidence from a clinical trial is to use a fixed statistical threshold of significance, also known as a P value or false positive rate, of 5% to determine whether a treatment is meaningful. Clinical trials with treatment effects that have a P value lower than 5% have traditionally been interpreted as statistically significant and those with higher P values are interpreted as insignificant. This 5% threshold is almost always applied across all diseases, but there’s reason to think that patients would be better served if the threshold were disease-specific.
For example, if we’re dealing with a disease that’s not particularly life-threatening, say acne medication, using a threshold of 5% for screening out false positives may be perfectly reasonable. But if we’re dealing with a potential therapy for a deadly disease like glioblastoma or pancreatic cancer where there’s currently no existing effective therapy and patients are likely to die anyway, patients might prefer to take more of a risk of a false positive so as not to miss a potentially effective therapy. We’ve developed a method for calculating the optimal threshold of false positives, given the severity of a disease and patient preferences regarding the costs and benefits of false positives versus false negatives. The FDA has been collaborating with us to construct a practical version of this framework.
EBO: What are you doing to get your ideas for financing healthcare put into practice?LO: I’ve advised a number of companies pro bono on how to apply these ideas and have been more directly involved in 2 start-ups, BridgeBio and Roivant, as a seed investor and a director. I’m now in the process of talking with a number of stakeholders about possible financing structures that can make gene therapies more widely accessible. The hope is that we can get something in place in time to be used as a template for the many other gene therapies that are coming online.
EBO: How enthusiastic are people about working with you and putting these ideas into practice?LO: I’m seeing tremendous enthusiasm from investors, drug manufacturers, and patients. There’s definitely money out there, waiting to be deployed and clearly a need for these new financing structures. Where we’re getting some hesitation—not so much pushback, but cautiousness—is among payers. State Medicaid plans don’t have an ability to amortize expenses; their budgets are set year to year and it’s not trivial for them to borrow funds or move expenses across budget years. As a result, they’re interested but more hesitant because they don’t really know what’s involved and they’ve got enough challenges to deal with currently. But given how many gene therapies are likely to be approved over the next few years, I suspect they’ll figure out a way to use these creative financing methods when the time comes.