Cost-Effectiveness, the Value of Immortality, and the Future of Gene Therapies
This past Thursday, the National Institute for Health and Care Excellence (NICE), published draft guidance recommending the United Kingdom’s National Health Service (NHS) not cover Bluebird Bio’s gene therapy Zynteglo. Zynteglo (betibeglogene autotemcel) “is the first and only one-time gene therapy for transfusion-dependent β-thalassaemia (TDT) that gives patients the potential to achieve transfusion independence.” This revolutionary treatment has the potential to cure patients’ disease and save them from a lifetime of blood transfusions and chronic anemia, among other health issues. The problem is it is EXPENSIVE, costing about ~$1.8M, broken into a 20% down payment and another 20% per year for each year of transfusion independence. Bluebird’s stock only fell about 2.1% on Thursday after news came out that NICE recommended against paying for the therapy, likely because the decision was highly predictable. At $1.8M, no therapy will ever fit NICE’s cost-effectiveness standards no matter how curative it is.
In fact, using NICE’s standards, the value of immortality would only be ~£857,143 GBP or ~$1,186,456. Why is immortality worth less than $1.2M? To figure this out, we need to understand how cost-effectiveness analysis works, and what NICE’s decisions mean for Bluebird and the entire gene therapy and rare disease industries.
There are two important metrics NICE uses, a cost per QALY and a discount rate. QALY stands for quality-adjusted life-year and is meant to represent a year of healthy life. The number of additional QALYs a medical treatment provides patients quantifies the benefit of treatment. Drugs and procedures can extend our lives as well as make life more enjoyable, hence the quality-adjusted aspect of a QALY. Besides longevity, we care about whether people are happy, pain-free, independent, etc.
Cost-effectiveness analysis looks at a metric called the Incremental Cost-Effectiveness Ratio or ICER. There are a few steps to calculating ICERs but in the simplest example let’s take two drugs: Drug A is a current standard of care treatment and Drug B is the next generation alternative. Drug B is both more costly and more effective than Drug A. To see if it’s worth the cost we calculate the ICER as the incremental cost divided by the incremental benefit or (CostB-CostA)/(QALYsB-QALYsA). Now we have a measure for the price of an additional QALY. This number is compared to a threshold value which is the upper limit of how much a system is willing to pay for a QALY. If it is above the threshold, the payor won’t cover the treatment, if it is below the threshold they generally will.
NICE generally uses a threshold of £20,000/QALY ($27,684), but they often will raise the number to £30,000 ($41,526) for “innovative” therapies or even higher for end of life treatments. In the United States, the Center for Medicare and Medicaid Services (CMS) has been banned from taking cost-effectiveness metrics like QALYs and ICERs into account, but academics in favor of QALYs tend to advocate for thresholds anywhere from $50–150k/QALY in the US. This is not the topic of today’s post, but I personally believe CMS should make coverage decisions based on a cost/QALY threshold equal to the GDP per capita.
An astute reader will now realize that having a multi-million-pound therapy be considered cost-effective is incredibly difficult. In the UK, for Zynteglo to be cost-effective at ~£1.3M, it would have to deliver ~66.67 QALYs. This would be a monumentally difficult hurdle for any therapy to cross. Virtually nothing can provide that much extra life to an individual. You would essentially need to cure a fatal disease in adolescents and provide them, fully-healthy lives going forward. However, while we could imagine a 67 QALY therapy, because of discounting, there can be no such thing.
Discounting says that a dollar (or effect) tomorrow is worth less than a dollar (or effect) today. This is due to a multitude of factors including a general time preference (we prefer things now to things later) and a level of uncertainty about the future (risk). In the US we tend to consider the 10-year treasury yield (currently at 1.2%) to be a risk-free rate which dictates the time preference for a guaranteed future payment. Other interest rates or stock returns are theoretically a function of the risk-free rate plus a risk premium. People, businesses, and even countries all have to account for discount rates when making decisions. As individuals, we face borrowing costs for our student loans, mortgages, and credit card debts. Discounting does not play a role in determining the cost-effectiveness of drug regimens where benefits occur at the same time as costs. However, when the timing of costs and benefits do not match up e.g. when you pay millions of dollars today for benefits years from now, discounting matters a lot.
Bluebird was hoping that NICE would use a 1.5% discount rate when evaluating Zynteglo, but the agency turned them down and instead used their standard 3.5% discount rate. This makes sense. According to statista, the risk-free rate of UK investments was 2.1% from 2015–2019. For a 1.5% discount rate to make sense, you would need an absolute guarantee that the drug would be 100% effective, and even then it’s an incredibly low number that only makes sense now in today’s low interest rate environment. However, while reasonable, sticking with a 3.5% discount rate is a death knell for gene therapy.
As I mentioned earlier, using NICE’s metrics of £30k/QALY and a 3.5% discount rate, the value of immortality would be £857,143. How do you get this? The formula for the present value (PV) of a perpetuity (constant stream of cash/benefits for an infinite number of years) is x/r where x is the yearly flow and r is the discount rate. Youthful immortality, where you never age, never get sick, and live forever would provide a value of 1 QALY per year for an infinite number of years. The present value of this QALY stream would be 1 QALY/.035 or 28.57 QALYs. In comparison, using a 1.5% discount rate would lead to a PV of 1 QALY/.015 = 66.67 QALYs. This decrease in the discount rate by 2 percentage points increases the present value by 133%, which is why Bluebird wanted their therapy to evaluated at a 1.5% discount rate, and why NICE’s decision is such a big deal.
So what’s the implication of NICE’s decision to use 30k/QALY and 3.5% discount rate? One thing is it means for gene therapies to gain NHS approval, costs will have to come down significantly. No level of efficacy, not even immortality, can justify £1M+ price tags. While every country makes its own coverage decisions, NICE is the leader people look to for cost-effectiveness analysis. If other nations follow the UK’s lead, none of the current or upcoming gene therapies in clinical trials will be included in national formularies. Now my personal view is big pharma and patient advocacy groups will invest heavily in pressure campaigns to prevent this from happening, but it nevertheless remains a possibility. What that would mean then is that gene therapies would become US-exclusive and already sky-high prices may go up even higher to justify the significant R&D and manufacturing expenses required to bring these treatments to patients. In the meantime, for therapies to gain global access and acceptance, the number one focus needs to be on reducing manufacturing costs. Unlike with traditional small molecule drugs, manufacturing gene therapies is incredibly expensive, costing hundreds of thousands of dollars. Refusal to pay for these innovations though creates a double-bind. Without investment today, prices will not decline in the future. Thus to reach a state where gene therapies become cost-effective in the future we likely have to overpay now. These overpayments or subsidies act as an investment in the technology supporting its growth and innovation until it can compete on its own, similar to how it took decades of government investment and subsidization for solar panels to generate electricity more cheaply than fossil fuels. From a policy perspective, that means the government should be focusing investment on research in lowering production costs and investing in technologies with the potential to scale.
Disclaimer: I am not an investment professional, and this is not meant to be taken as investment advice. I am currently working for a competitor to BlueBird bio also developing gene therapies for rare diseases.