Investigational Medical Products for Rare Diseases: Ethical Concerns and Unresolved Issues with Expanded Access

By Alison Bateman-House, PhD, MPH, MA and Lisa Kearns, MS, MA

Introduction

Patients who have no therapeutic or diagnostic options available may seek access to investigational (experimental) products in development by enrolling in clinical trials. For seriously ill patients who cannot join a trial, Expanded Access (EA) may be a viable alternative. EA is a United States Food and Drug Administration (US FDA) regulatory pathway by which certain patients may access unapproved medical products outside of trials, if the entities developing the products agree to provide them. EA can take on heightened importance for rare- and ultra-rare disease communities, for whom there are often limited approved drugs/diagnostics and trial options. Yet EA––as an exceptional therapeutic effort involving products of unknown efficacy and safety––raises myriad ethical concerns, chief among them how to balance the welfare of individual patients with that of a patient population; whether access to products via EA can ever be equitable; and how to navigate the challenges EA presents regarding informed decision-making. In this article, we discuss these concerns in the context of EA for rare diseases, fact-check some common perceptions, and discuss unresolved issues that, while thorny, demand deliberation and eventual resolution.

The Purpose of Expanded Access

The majority of patients who try investigational products, defined here as medical products in development prior to approval by a regulator for any use, do so in clinical trials. Patients often enroll in trials in hopes of benefiting personally; however, they can also be motivated by a desire to help generate knowledge for the benefit of other patients with their condition. Those who participate in trials help advance understanding of a condition and its treatment by contributing to the collection of safety and efficacy data. This societal and public health gain is why most nations prioritize participation in clinical trials over non-trial access to investigational products. Still, many nations allow non-trial access to some patients who cannot participate in a trial; such access occurs under a dizzying array of terminology including “compassionate use,” “expanded access,” “preapproval access,” “special access,” “managed access,” “single patient INDs,” “named patient programs,” and more [1].

Non-trial access

regulations vary globally but generally require eligible patients to have serious or life-threatening conditions with no untried approved therapeutic options. The potential benefit of the experimental product must outweigh its risk of harm, and a patient (or surrogate decision-maker) must give informed consent. Additionally, eligible patients must be unable to participate in a clinical trial, and the product’s developer (“sponsor”) must have determined that provision of the product will not interfere with its clinical development, such as by making trial accrual less appealing [2]. These stipulations are designed to prioritize the clinical development of the product while still providing seriously ill patients non-trial access to potentially beneficial treatments or diagnostics.

In the US, non-trial access to investigational medical products is permitted via two routes: EA and Right to Try (RTT). Importantly, access via either route hinges on the willingness of the sponsor, most often a biopharmaceutical company, to provide the requested product. EA, the far more common and preferred route, accommodates both individuals and groups of patients and allows access to products at any stage of development. In single patient EA, the patient’s physician creates a customized plan in cooperation with the sponsor, the FDA, and an ethics entity called an institutional review board (IRB). Figure 1 reflects the most recent data on single patient EA requests from FDA’s Centers for Drug Evaluation and Research (CDER) and Biologics Evaluation and Research (CBER). Note that the FDA allowed the overwhelming majority of single patient EA requests to proceed.

There is also a second type of EA, for larger groups of patients. These “cohort” EA programs resemble clinical trials in that they operate under protocols developed by a sponsor and approved by the FDA and an IRB. However, while in clinical trials inclusion and exclusion criteria are established to prevent harm to participants and to facilitate the generation of quality data from which robust conclusions can be derived, in a cohort EA program these criteria are set to prevent harm to patients and to ensure that participants are truly ineligible for a clinical trial. EA programs can be “intermediate” size, accommodating two to an unspecified number of patients, or “treatment” size, accommodating large numbers of patients. Treatment EA programs are frequently opened after sponsors have submitted marketing applications for the investigational products and are intended to provide wide-scale access to patients who cannot wait for commercial availability of the product. RTT is an alternative route to EA requiring neither FDA nor IRB oversight. Only single patients who suffer from life-threatening conditions are eligible, and it applies only to drugs, which must be in active development and have completed a successful Phase I trial. The many ethical concerns surrounding RTT have been addressed extensively elsewhere [3,4]. Because this pathway is used exceedingly rarely, we do not discuss it further.

Expanded Access Perceptions and Facts

An entire journal could be devoted to clarifying what EA is meant to achieve, how it works, its legal history, and more. The same goes for ethical concerns and unresolved issues: the below is by no means exhaustive. The items we’ve included are ones we have encountered most often in years of studying the topic.

Perception: It is difficult to access medical products via EA, and this is due to the FDA or IRB.

Reality: The FDA allows virtually all EA requests to proceed (see Figure 1), as do IRBs. [5] However, these entities can only review EA requests submitted to them, and they receive requests only after a sponsor has agreed to grant access; if a sponsor says no, the request will never be reviewed by the FDA or an IRB.

The primary gatekeeper to EA is the sponsor, which has no legal obligation to allow preapproval access to its products. How often sponsors deny requests for EA is a mystery: some major pharmaceutical companies voluntarily report the numbers of single patient requests they receive and grant, but there is no public accounting by the majority of sponsors. Many medical products are initially developed by small companies, with larger companies acquiring the rights to the products if they show commercial potential. Products targeting rare diseases often have less commercial appeal than those in development for much larger patient populations; thus, rare disease products tend to be brought to market by smaller companies. This has implications for EA, as, anecdotally, small companies are less likely to allow EA than large ones, for a variety of reasons [6].

Perception: EA requires time and effort on the part of physicians and patients.

Reality: Just as sponsors have no legal obligation to grant access to their products via EA, physicians have no legal obligation to facilitate EA, and some patients will have to devote additional time and expense to find physicians willing to help them.

Cohort EA programs, which are often established by sponsors but in some cases can be initiated by physicians, resemble clinical trials and thus require significant effort to establish and monitor. Clinicians who take leadership roles in these efforts will spend considerable time communicating with the FDA, IRB, sponsor, and others. Clinicians without leadership roles need only to refer patients to the EA program, as they would do for a clinical trial.

For single patient EA, which involves a physician, sponsor, IRB, and FDA, the majority of effort falls to the physician, who must shepherd the entire process, often without dedicated time, remuneration, support staff, or knowledge or training about how to do so [7]. Recent years have seen several efforts to simplify single patient EA and make navigating the process more straightforward. In 2019, the FDA launched Project Facilitate, a concierge program that guides oncologists through the process[8]. The Reagan-Udall Foundation for the FDA and individual biopharmaceutical companies have worked to enhance transparency of EA-related policies and procedures in accordance with the 21st Century Cures Act, which mandates disclosure of certain sponsor’s EA policies [9]. Despite these laudable steps, there are still reasons to believe that physicians do not understand EA very well [7,10], which can hinder their ability to navigate the process. Patients with rare diseases may actually fare better on this front, because tightly knit networks can help increase awareness and understanding of EA for both patients and clinicians.

Perception: Sponsors do not grant EA requests because of cost.

Reality: Some large companies with lots of products on the market are willing to provide EA regardless of its expense. This may not be the case for many companies in the rare disease space. Despite inducements such as those created in the Orphan Drug Act [11], sponsors can be hesitant to tackle rarer diseases, and when they to do, it may be with less capital and at higher risk than for sponsors developing medical products for more prevalent conditions. This can lead sponsors to be more conservative in responding to EA requests, by either declining to offer EA or limiting the size of programs. While federal regulations allow sponsors to recoup direct manufacturing costs, sponsors usually provide products via EA for free.

Sponsors have other concerns beyond cost. One is that serious adverse events that occur in an EA context may threaten the development of the product. Although this concern is overblown—FDA has said that it realizes that EA involves very sick patients and that not all serious adverse events necessarily stem from the experimental intervention [12]—until sponsors internalize this, fears of regulatory, and potentially investor, repercussions will persist, perhaps leading sponsors to be leery of EA. Furthermore, sponsors developing single-administration therapeutics for very small populations may worry that EA will reduce their eventual consumer markets. For example, a sponsor that provides an investigational gene therapy  via EA will lose that patient from its already limited pool of potential consumers post-approval. Finally, sponsors may decide not to offer EA due to concern about how to most ethically allocate a limited supply of a product.

Perception: EA requires time and effort on the part of physicians and patients.

Reality: Just as sponsors have no legal obligation to grant access to their products via EA, physicians have no legal obligation to facilitate EA, and some patients will have to devote additional time and expense to find physicians willing to help them.

Cohort EA programs, which are often established by sponsors but in some cases can be initiated by physicians, resemble clinical trials and thus require significant effort to establish and monitor. Clinicians who take leadership roles in these efforts will spend considerable time communicating with the FDA, IRB, sponsor, and others. Clinicians without leadership roles need only to refer patients to the EA program, as they would do for a clinical trial.

For single patient EA, which involves a physician, sponsor, IRB, and FDA, the majority of effort falls to the physician, who must shepherd the entire process, often without dedicated time, remuneration, support staff, or knowledge or training about how to do so [7]. Recent years have seen several efforts to simplify single patient EA and make navigating the process more straightforward. In 2019, the FDA launched Project Facilitate, a concierge program that guides oncologists through the process [8]. The Reagan-Udall Foundation for the FDA and individual biopharmaceutical companies have worked to enhance transparency of EA-related policies and procedures in accordance with the 21st Century Cures Act, which mandates disclosure of certain sponsor’s EA policies [9]. Despite these laudable steps, there are still reasons to believe that physicians do not understand EA very well[7,10], which can hinder their ability to navigate the process. Patients with rare diseases may actually fare better on this front, because tightly knit networks can help increase awareness and understanding of EA for both patients and clinicians.

Perception: Sponsors do not grant EA requests because of cost.

Reality: Some large companies with lots of products on the market are willing to provide EA regardless of its expense. This may not be the case for many companies in the rare disease space. Despite inducements such as those created in the Orphan Drug Act [11], sponsors can be hesitant to tackle rarer diseases, and when they to do, it may be with less capital and at higher risk than for sponsors developing medical products for more prevalent conditions. This can lead sponsors to be more conservative in responding to EA requests, by either declining to offer EA or limiting the size of programs. While federal regulations allow sponsors to recoup direct manufacturing costs, sponsors usually provide products via EA for free.

Sponsors have other concerns beyond cost. One is that serious adverse events that occur in an EA context may threaten the development of the product. Although this concern is overblown—FDA has said that it realizes that EA involves very sick patients and that not all serious adverse events necessarily stem from the experimental intervention [12]—until sponsors internalize this, fears of regulatory, and potentially investor, repercussions will persist, perhaps leading sponsors to be leery of EA. Furthermore, sponsors developing single-administration therapeutics for very small populations may worry that EA will reduce their eventual consumer markets. For example, a sponsor that provides an investigational gene therapy via EA will lose that patient from its already limited pool of potential consumers post-approval. Finally, sponsors may decide not to offer EA due to concern about how to most ethically allocate a limited supply of a product.

Expanded Access for Rare Diseases: Ethical Concerns

Three-quarters of rare diseases affect children, and often these conditions are fatal and/or lead to significant morbidities. Despite the undeniable appeal of preapproval access for this special class of patients, EA nevertheless raises troubling ethical concerns. We and colleagues have argued that developers of gene therapies have an ethical obligation to consider providing EA to their products [13], even though such access is not legally required. We premise this argument on the fact that many gene therapy products address serious unmet needs that require urgent intervention and thus merit proactive deliberation about foreseeable EA requests. This argument extends to non-genetic interventions for serious indications that require urgent intervention.

EA policies should reflect the severity of the medical condition, the available alternatives, the product’s known and unknown risks and benefits, the current and future availability of the product, and the modality of administration. Of paramount importance when sponsors make EA policies is that there should be considerations about informed consent and equity of access.

Informed Consent

Short windows for commencing treatment, few alternative options, lack of medical literacy, and hype about interventions’ purported benefits can all lead patients or guardians to overestimate an unproven treatment’s therapeutic benefit and underestimate its risk. In addition to a thorough explication of known and unknown risks and benefits, the informed consent process should cover relevant financial and other concerns; for example, that use of the invention may preclude the patient from receiving other medical products, in or out of a trial. Ethical concerns about patients’ ability to provide valid informed consent to unproven interventions are not particular to EA nor to rare disease. However, this already troublesome issue is intensified with rare diseases, because of the scarcity of approved treatments and trials—and thus scarcity of data about risks and benefits—and because so many rare diseases strike children, meaning decisions about whether to attempt EA will be made by parents or other guardians on behalf of patients unable to make their own treatment decisions. Regulators, clinicians, and ethicists, among others, have long sought to address the difficulty in making treatment decisions for children, and regulations and policy reflect this. However, EA undergoes less strict ethical review than do clinical trials, so clinicians and patient advocacy groups must play a larger role in ensuring patients or surrogates understand all aspects of their decision.

Justice

Inequitable access to healthcare in the US and around the world includes inequitable access to investigational medical products. As a pathway, EA is fundamentally limited by a patient’s ability to find a physician who is willing and has the resources to identify a possible intervention and then pursue access to it. This factor has a hidden socioeconomic component, as those patients who are barred by lack of financial and other resources (the inability to find childcare or take time off from work/school, for instance) are less likely to be able to search for a willing physician, to seek second opinions, or to travel for treatment. This is particularly relevant for patients with rare diseases who engage in “diagnostic odysseys,” expensive and time-consuming searches for accurate diagnoses; even if they obtain correct diagnoses, there may be relatively few specialists or institutions that can provide appropriate treatment. Uneven access to EA on socioeconomic grounds worsens if patients must pay out of pocket for either the unapproved product itself or ancillary costs, such as physician fees.

Public and private payers typically do not cover unapproved treatments and may refuse to pay associated costs as well, leaving even those with insurance potentially vulnerable to large expenses [14]. In sum, EA is typically a viable option only for those with the ability, time, and resources to find and afford a willing physician and promising intervention.

The inequity in EA is brought into even starker relief in a particular niche of the rare disease world: the development of individualized therapeutics. The most famous such case to date is that of milasen, an antisense oligonucleotide developed for one patient, a child with a unique mutation of Batten disease [15]. Estimated costs of the development and administration of milasen are estimated to be seven figures, not including costs incurred by the family for travel and housing for treatment, and the like [16].

With individualized therapeutics, there is no intention to develop the product for regulatory approval, so EA is not something that happens secondary to a clinical trial of an experimental product. Rather, non-trial preapproval use of the drug via EA is the intended, and only, use. Yet, EA was meant to be an exceptional option accompanying clinical development of a medical product. This major shift in usage of the pathway will require policymakers and regulatory bodies, including IRBs, to reconsider the process and their roles in it [17]. Likewise, it will require grappling with the principle of justice, given that not all patients have the resources to self-fund such endeavors, and not all patients will be able to garner charitable support [18,19].

Unresolved Issues in Expanded Access for Rare Diseases Cost

As mentioned, cost can be an important factor for sponsors deciding whether to offer EA. Cost is a bigger consideration if the product in question entails complicated manufacturing or delivery or a companion diagnostic. Autologous treatments, such as CAR-T or stem cell–based products, involve collection of biological materials from a patient, customized development of a product, and then delivery of that product to the patient, possibly via surgery—associated costs that may fall to the patient. Although the hope is that ongoing development of individualized therapeutics will lead to techniques that can eventually reduce costs, it’s difficult to imagine individualized medicine becoming easily affordable, especially if insurers continue to view it as experimental. Unfortunately, inequities in access will only persist and deepen. This issue must be factored into all decisions about EA, and the status quo must be recognized as unacceptable by anyone concerned about justice in access to healthcare.

Unintended Consequences of Early Intervention

For many rare diseases, early intervention is vital to halt or slow disease progression and forestall reduced quality of life and the possibility of early death. This has two implications. The first is that there must be ways to detect diseases in time for early intervention. This requires infrastructure for early detection— infrastructure that must be designed with equity in mind. For instance, universal childhood testing for a rare diseases would be costly, yet through an equity lens, it would be morally preferable to the current system, in which often only those who are able to seek out and pay for such testing are able to benefit. Allocating resources to one program, such as universal newborn screening for diseases, typically means that those resources will have to be diverted from other programs, so there are no easy answers here.

Another implication of the drive for early intervention is that making products available via EA may have the unintentional downstream impact of depleting both the pool of patients available for future clinical trials and the future market for a product. For both situations, providing broad non-trial access to patients who may benefit from it may result in harm to a wider patient community, which relies upon both the rapid conduct of trials to advance knowledge and ongoing investment, most frequently by profit-driven biopharmaceutical companies, to create next generation interventions. The most obvious solution to the depletion of future trials in the rare disease space is for sponsors working on a common therapeutic target to be transparent about their plans for current and future clinical trials and EA programs and then work together—pre-competitively—to ensure that a limited pool of patients is preferentially channeled into trials and those who cannot be accommodated in a trial would have the opportunity for EA. This goes hand in hand with championing innovation in clinical trial design so that more patients can access experimental products through them, thereby reducing overall demand for EA. A silver lining of the coronavirus pandemic is that it has forced researchers to think creatively about how to conduct trials remotely, knowledge that should be applied to post-pandemic research.

Conclusion

Everyone involved with rare diseases must understand what EA is intended to do and the ethical and other challenges it entails. The valuable time and energy that have been wasted on initiatives—such as Right to Try legislation, which was premised on the false claim that the FDA was a near-insurmountable bureaucratic obstacle to EA—would have been better spent tackling real issues concerning the use of experimental medical products for rare diseases, such as what to do about the fact that trials are often burdensome for patients and can limit participation and cost is a significant barrier. Admittedly, medical innovation is expensive and involves a complex web of stakeholders with some shared and some unique interests. This makes it difficult to state hard and fast rules about how or when access to unapproved medical products should occur or what changes to current practices are needed. This difficulty only increases when the interventions are novel or complex and for indications that face relatively steeper market and diagnostic hurdles. Increased understanding of these issues and a streamlined focus on the underlying ethical and other concerns will help to expedite progress for the benefit of patients and the general public.

Alison Bateman-House, PhD, MPH, MA ([email protected]), Division of Medical Ethics, Department of Population Health, NYU Grossman School of Medicine, New York.

Lisa Kearns, MS, MA ([email protected]), Division of Medical Ethics, Department of Population Health, NYU Grossman School of Medicine, New York.

References

1. Kimberly LL, Beuttler MM, Shen M, Caplan AL, Bateman-House A. (2017). Pre-approval access terminology: a cause for confusion and a danger to patients. Ther Innov Regul Sci. Jul;51(4):494-500. doi: 10.1177/2168479017696267. Epub 2017 Mar 3. PMID: 30227050.
2. Bunnik EM, Aarts N, van de Vathorst S. (2018). Little to lose and no other options: ethical issues in efforts to facilitate expanded access to investigational drugs. Health Policy. Sep;122(9):977-983. doi: 10.1016/j.healthpol.2018.06.005. Epub 2018 Jun 18. PMID: 29935731.
3. Folkers K, Chapman C, Redman B. (2019) Federal Right to Try: where is it going? Hastings Cent Rep. Mar;49(2):26-36. doi: 10.1002/hast.990. PMID: 3099828.1
4. Bateman-House A, Kimberly L, Redman B, Dubler N, Caplan A. Right-to-try laws: hope, hype, and unintended consequences. (2015) Ann Intern Med. Nov 17;163(10):796-7. doi: 10.7326/M15-0148. Epub 2015 Sep 29. PMID: 26413841.
5. Chapman CR, Shearston JA, Folkers KM, Redman BK, Caplan A, Bateman-House A. Single-patient expanded access requests: IRB professionals’ experiences and perspectives. AJOB Empir Bioeth. 2019 Apr-Jun;10(2):88-99. doi: 10.1080/23294515.2019.1577192. Epub 2019 Apr 9. PMID: 30964737.
6. Chapman CR, Moch KI, McFadyen A, Kearns L, Watson T, Furlong P, Bateman-House A. (2019). What compassionate use means for gene therapies. Nat Biotechnol. Apr;37(4):352-355. doi: 10.1038/s41587-019-0081-7.
7. Chapman CR, Belli H, Leach D, Shah L, Bateman-House A. (2021). A survey of pediatric hematologists/oncologists’ perspectives on single patient Expanded Access and Right to Try. Med Access Point Care. 5. 239920262110059. 10.1177/23992026211005991.
8. Project Facilitate. Accessed Sep 3, 2021.
9. Kang S, Chang S, Ross JS, Miller JE. 2021. Implementation of 21st Century Cures Act expanded access policies requirements. Clin Pharmacol Ther. Aug 24. doi:10.1002/cpt.2401. Online ahead of print.
10. Moerdler S, Zhang L, Gerasimov E, Zhu C, Wolinsky T, Roth M, Goodman N, Weiser DA. (2019). Physician perspectives on compassionate use in pediatric oncology. Pediatr Blood Cancer. Mar;66(3):e27545. doi: 10.1002/pbc.27545. Epub 2018 Nov 8. PMID: 30408307.
11. 97th Congress (1981-1983). Public Law 97-414. Accessed Sep 13, 2021.
12. U.S. Food and Drug Administration. (2016, updated 2017). Expanded access to investigational drugs for treatment use—questions and answers: guidance for industry. Accessed Aug 28, 2021.
13. Kearns L, Chapman CR, Moch KI, Caplan AL, Watson T, McFadyen A, Furlong P, Bateman-House A. (2021). Gene therapy companies have an ethical obligation to develop expanded access policies. Mol Ther. Apr 7;29(4):1367-1369. doi: 10.1016/j.ymthe.2021.03.008. Epub 2021 Mar 13. PMID: 33714373; PMCID: PMC8058481.
14. McBride Folkers K, Bateman-House A, Robertson C. (2020). Wake Forest Law Rev. Vol11:1.
15. Kim J, C Hu, C Moufawad El Achkar, LE Black, J Douville, A Larson, MK Pendergast,SF Goldkind, AL Eunjung, et al. (2019). Patient-customized oligonucleotide therapy for a rare genetic disease. N Engl J Med 381(17):1644-1652.
16. Hayden EC. This girl’s dramatic story shows hyper-personalized medicine is possible—and costly. Accessed Sep 4, 2021.
17. Bateman-House A, Kearns L.(forthcoming) The Ethics of N=1 Development for Rare Diseases. Nucleic Acid Ther.
18. Snyder J, Bateman-House A, Turner L. Is right to try being tried? Using crowdfunding data to better understand usage of nontrial pre-approval access pathways. (2020). Regen Med. Aug;15(8):1979-1985. doi: 10.2217/rme-2020-0043. Epub 2020 Oct 7. PMID: 33023369.
19. Kenworthy NJ. Crowdfunding and global health disparities: an exploratory conceptual and empirical analysis. (2019) Global Health. Nov 28;15(Suppl 1):71. doi: 10.1186/s12992-019-0519-1. PMID: 31775896; PMCID: PMC6882318.