The next generation of rare disease drug policy: ensuring both innovation and affordability

To target research, development and commercialization of drugs to treat ultra-rare conditions requires agreement on a definition distinguishing ultra-rare disorders from other rare diseases. One reasonable threshold, following European approaches, would be to identify as ultra-rare those conditions with fewer than 10,000 patients in the USA. Most stakeholders agree that even with high prices, products for patient populations this small are very hard to commercialize [17]. Some observers have argued that population standards should consider not just the USA but the global patient population in order not to overapply special incentives where they are not needed.

This research found that patient advocates for rare conditions have not favored the idea of separately identifying a narrower ultra-rare group. They argue that even if ultra-rare conditions receive a boost to support innovation, such a separation over time would give policymakers too much leeway to decrease incentives that are still vital for future innovation across the broader range of orphan conditions. If such a distinction is to be drawn, however, advocates underscore the importance of considering not only absolute patient numbers but also patient-level quality-of-life indicators, such as disease severity, level of disability and premature mortality. They argue that these factors should also be considered if a subset of orphan conditions is to be carved out and given special incentives. Policymakers, however, would need to weigh the potential advantages of including additional, less objective criteria, with a subjective algorithm for designating which conditions should be eligible for additional incentives.

Unless development of ultra-rare treatments is taken over by government-directed programs, enhanced financial incentives would be needed to stimulate more research and development of these treatments. One option is to increase ODA tax credits or other subsidies specifically for products treating ultra-rare disorders. Such a change would explicitly prioritize ultra-rare diseases as requiring more incentives than products for other rare conditions, though some stakeholders worry this could create too strong an incentive to move resources toward conditions that impact fewer people. Regardless, changes to the tax credits may be insufficient to overcome the market challenges faced by ultra-orphan drugs and enhancements to the tax credit may not be immediately useful for early-stage companies because of the structure of a tax credit.

Another, more direct approach to supporting the development of ultra-rare drugs would be to expand direct federal funding for and involvement in ultra-rare drug research and development. One option would be to establish a new federal authority to conduct critical research and development for ultra-rare conditions, akin to what the Defense Advanced Research Projects Agency does for technology and military innovation. The Defense Advanced Research Projects Agency does not have its own laboratories or research facilities, instead directing promising research pathways through grant making and partnerships with scientists. In a similar model, the Bespoke Gene Therapy Consortium is an example of a public-private partnership among the NIH, FDA, industry and nonprofit partners that seeks to foster development of gene therapies by developing tools to streamline the development process [18]. Operation Warp Speed, which accelerated the COVID-19 vaccine development, is another potential model. In that example, the federal government directly subsidized research and development costs, contracted to support manufacturing capacity, purchased necessary ingredients and supplies and provided advance-purchase commitments to underwrite market risk [19]. The federal government could also encourage ultra-rare drug development through increasing NIH funding for basic research in ultra-rare conditions, creating a loan or other system to offset manufacturing expenses, establishing subsidies to underwrite commercialization costs and guaranteeing market access via pre-established coverage and reimbursement expectations in Medicare and Medicaid.

Looking abroad, another option for creating special arrangements for ultra-orphan drugs has been to create special assessment pathways and reimbursement mechanisms at the national level. For example, NICE in the UK has a separate pathway for assessment of ‘Highly Specialized Technologies’ that in principle is willing to accept much higher cost–effectiveness thresholds for treatments for ultra-rare conditions [20]. Congress could consider creating a similar national coverage structure for ultra-rare treatments through the Centers for Medicare & Medicaid Services that would leverage the existing coverage with evidence development mechanism to accelerate access to these treatments with a corresponding evidence development requirement. This mechanism could also be linked with a requirement for value-based pricing (see below), at least during the evidence generation phase of coverage. If this approach, with or without a pricing component, were limited expressly to treatments for ultra-rare conditions, the resulting improvement in access might spur additional interest in developing these treatments while creating manageable increases in spending.

Once a product to treat an ultra-rare condition is brought to market, whether developed by industry or government, its price determines its cost–effectiveness. If the product is developed by government, pricing can be done by administrative fiat. If it is developed or codeveloped by industry, the product's price, shaped as described above, underpins the economic incentives for innovation. As such, prices paid by payers for products that treat ultra-rare conditions, especially those with the lowest prevalence, may need to exceed typical thresholds used in cost–effectiveness assessments.

Instead of relying on traditional cost–effectiveness assessments, prices for ultra-rare products could be developed based on a different pricing paradigm, according to which the rates of return for investments in developing orphan drugs should not be greater than the industry average [21]. In this approach, a calculation would be done of the maximum allowable price that society should be willing to pay. Such a change would represent a major departure from current US reimbursement models in which the government does not set prices. A downside of this approach is that it could reward inefficient research and development programs while under incentivizing the kinds of risk taking needed on early assets that promise substantial clinical gains. Nonetheless, a rate-of-return approach could be twinned with cost–effectiveness, with the greater price calculated through either paradigm the one that would be accepted by payers. Ultimately, by managing the balance between innovation and affordability more explicitly and consistently, a reasonable rate-of-return methodology could generate pricing recommendations that would do a better job of balancing the need for greater incentives for ultra-rare treatments within a more predictable and affordable framework.

Many orphan products also secure FDA approval for nonorphan indications, products sometimes referred to as ‘partial orphans’. As of 2018, 23% of approved orphan drugs also had non-orphan indications [22]. While ODA exclusivity benefits apply only to orphan indications, once a product receives approval for an expanded indication, manufacturers almost always maintain the initial high ‘orphan-level’ pricing while expanding their product sales and revenue [23]. Thus, it is perhaps not surprising that partial orphans are particularly common among drugs with the highest overall revenue. In 2018, five of the six top-selling drugs in the USA were partial orphan drugs, and a recent Office of Inspector General report found that a majority of the highest-expenditure drugs in Medicare have been granted at least one orphan designation [24,25]. As one extreme example, pegfilgrastim (Neulasta^®) had the fourth highest spend among partial orphan drugs currently on the market, but only 0.6% of this amount was for its orphan indication of acute radiation syndrome [24].

There are clear benefits for manufacturers in testing the viability of promising pharmaceutical agents first within narrowly targeted patient populations and if found successful, to expand development more broadly to additional indications. As such, it is neither surprising nor problematic that some products launch as an orphan drug before gaining nonorphan indications. However, the evident commercial success of partial orphans, in part derived from their sustained ‘orphan pricing’ as patient populations expand, suggests that ongoing federal research and development tax credits and waived user fees may not be necessary or appropriate. Specific policy options for modulating orphan drug benefits for partial orphans are described below.

One potential policy reform is to limit or remove orphan drug incentives once a product is approved for nonorphan indications or once it exceeds a given threshold of revenue (e.g., US$200 million). Such a change would tend to focus orphan drug incentives on products that need it most – those treating very small patient populations and/or those for which competition has limited their market share. The potential risk in this approach is that a reduction in ODA benefits tied to revenue might discourage companies from taking the risk of conducting additional clinical trials to assess the effectiveness of a product for a broader population, thus limiting the ultimate clinical benefit to the broader population of some treatments that start out as orphan drugs.

Another option to rebalance incentives for partial orphan drugs would be to reassess and sharpen the FDA's definition of distinct diseases. Increased use of new molecular biomarkers to diagnose and treat conditions is resulting in narrower definitions of distinct diseases and thus some heterogeneous conditions that did not qualify as rare are now being subdivided into multiple ‘new’ rare diseases. As this trend continues, the FDA could change its approach to defining distinct diseases to favor the preservation of broader indications. A broader definition would reduce the incentive for manufacturers to start their development program targeting a very narrow population, potentially blunting the trend of leveraging early orphan pricing into larger populations over time. As with any narrowing or reduction of ODA benefits, however, this policy option might increase uncertainty regarding the scientific and commercial success of some emerging treatments, producing less investment and innovation in new treatments for rare diseases.

Policymakers could restrict the 340B exemption for Affordable Care Act-expanded covered entities to apply to only the utilization of an orphan indication, rather than all indications the partial orphan is designated for, which would dramatically reduce the financial incentives for partial orphan products [24]. This change would likely require legislation to surmount the US District Court decision that struck down the Health Resources and Services Administration's interpretive rule. However, as with all potential reductions in ODA benefits, this policy change harbors some risk of driving investment and innovation away from certain rare disease areas and it would be challenging to implement, as it would require participating providers to document, monitor and report their 340B drug utilization by indication. While such reporting is conceptually possible in the context of well-integrated electronic medical records and data systems, in practice it could create a significant operational burden on 340B entities.

Federal investment is needed to spur the development of more robust systems for capturing and analyzing observational data to meet the needs of patients, clinicians and payers. These data systems should be developed so that they can capture patient-reported outcomes reflecting broader patient and family effects of treatment. Federal funding, along with help from disease and patient groups, can accelerate efforts to improve evidence generation by sponsoring rare condition registries. Broadening rules on data ownership and using federally funded program claims data (e.g., Medicare and Medicaid) to augment registry data that could further enhance these resources. The federal government should consider additional strategies to expand registry development for rare conditions, including supporting their development by patient and disease groups and by requiring registry participation by entities delivering services to Medicare or Medicaid patients with the associated rare condition.

Increased spending on orphan drugs has paralleled the growth in orphan drug development. Spending for orphan indications increased from 2% of invoice spending on drugs in 1992 to 11% in 2019 [5]. An additional 16% of drug spending was on nonorphan indications of drugs that also had orphan indications [5]. Such spending increases are driven by increases in the volume of orphan drugs and because these products have higher prices. In 2017, the average annual cost of orphan drugs at launch was 25-times higher than the annual cost of treatment for non-orphan drugs [26]. Another analysis found that among the top 100 drugs by US sales, the average cost of treatment for orphan drugs was 4.5-times that of nonorphan drugs (US$150,854 vs $33,654 per year), a difference of US$117,200 on average [27].

In the absence of effective and affordable mechanisms to manage spending for extremely high-cost, low-utilization products, some plan sponsors have resorted to the extreme of stripping out cell and gene therapies entirely from their health benefits [28]. Health plan and pharmacy benefit manager executives interviewed for this report indicated that they experience regular and growing pressure from self-insured employers to exclude coverage for these products and anticipate that this trend will only increase as the numbers of these treatments add to the actuarial impact of pharmaceutical coverage.

As noted earlier in regard to several policy options, value-based price regulation could be a complement to other policies seeking to realign incentives and improve affordability. There are numerous mechanisms available, including the use of international reference pricing, or pricing based on cost–effectiveness thresholds and algorithms suggested by value assessment by groups such as the Institute for Clinical and Economic Review (ICER) [37]. One benefit of this approach is that the value-based price-setting methodology might shift over time as additional evidence is generated. Further, value-based price setting creates incentives for investment in evidence development to demonstrate the clinical benefits of emerging treatments and provides handsome market rewards for products bringing significant added benefits to patients while scaling those rewards down for drugs that do little to improve patient outcomes.

In adopting a value-based price-setting mechanism for orphan products, policymakers must consider whether products that treat rare and ultra-rare conditions should be afforded different standards for cost–effectiveness. ICER addressed this question as part of its consideration of adapted assessment methods for treatments for ultra-rare disorders [17]. While the custom of accepting higher prices for ultra-rare disease products suggests a societal willingness to pay more for these products, ICER concluded that the logic and ethics of opportunity costs suggested that cost–effectiveness thresholds should not shift systematically solely on the basis of rarity and that such shifts threaten the goals of health equity. However, absent a comprehensive approach to encourage affordable pricing for rare diseases while encouraging investment in ultra-rare treatments, ICER could consider adjusting its cost–effectiveness thresholds (or granting higher weights to health gains for patients with ultra-rare disorders) to achieve the policy aim.

As with any policy that would lead to lower prices for some orphan products, value-based pricing mechanisms may result in pushing prices too low to incentivize investment in areas in which success is less certain, or in which the clinical gains from treatment would be relatively small. Some would argue that progress in the treatment of orphan diseases often seems to come in small steps and that prices generated by value-based mechanisms would not be adequate to keep the field moving forward. The contrasting argument is that patient populations between 10,000 and 200,000 appear large enough to generate sufficient revenue to keep investments and risk taking at robust levels. For example, a product that treated only 10,000 patients at a price of US$100,000 per year (22% of treated patients today receive an orphan drug that costs US$100,000 or more [5]) would generate an annual revenue stream of US$1 billion for the manufacturer. Nonetheless, judging the impact of value-based prices on the broad range of orphan drug development remains a hypothetical exercise, with the potential benefits and negative consequences frequently debated.

Insurance structures depend on the presence of shared risk pools large enough that diseases are distributed relatively evenly across the risk pools. For common conditions, with reasonably modest costs, risk pools can be small, down to the level of the individual self-insured employer and its employees. But for rare and ultra-rare conditions with extremely high costs, risk pools need to be very large – otherwise, the individual employer or small group of plan sponsors might not be able to absorb the unexpected cost shock of an unanticipated number of patients with rare and expensive conditions.

Many national health plans are now trying to create insurance products to manage the extreme risk of unanticipated high-cost orphan drug treatments. Some federal policymakers have supported this approach [38]. One example of these efforts is Embarc Benefit Protection, offered by Cigna. The program carves out coverage for cell and gene therapies into a set per-member, per-month premium for employers and other plan sponsors who join the network. This premium structure protects the individual employer from the financial burden of an unanticipated high-cost therapy while allowing individual patients to access therapies without any out-of-pocket payment [39].

Although the commercial experience with these carve-out programs is limited, policymakers may wish to consider launching one at even larger scale at the national level. A new national benefit for cell and gene therapies for orphan diseases would create a single national risk pool for all identified rare conditions. Such a program would carve out payment for orphan products generally, or cell and gene therapies specifically, from other public or private coverage. Carve-outs and national risk pools have been proposed in several forms by some policymaking bodies, including the Medicaid and CHIP Payment and Access Commission [40]. Others have proposed carving these products out of Medicare's hospital payment system to ensure adequate reimbursement [41].

A carve-out could include a single program (e.g., Medicaid) or could aggregate multiple public or private purchasers via optional or mandatory participation. In its most expansive form, a carve-out program would constitute a new federal entitlement for all Americans. While more limited carve-outs would be more politically feasible in the near term, fragmenting a diffuse risk pool could create unintended consequences that result in some payers seeking to deny coverage for these products in order to drive coverage into the carve-out.

A standardized national benefit for rare disease products generally, or cell and gene therapies specifically, could create uniform coverage and risk-pooling arrangements across the patient population, giving payers the opportunity to compete on providing patients with fair and equitable access to products and appropriate care and home support services. Pricing could be based on an outcomes-based contract or a value-based pricing arrangement, as described above. Such a proposal would be a significant change to current insurance coverage that would have widespread effects on stakeholders.