Patients with rare diseases have had difficulty obtaining disease-specific treatments due to inadequate biotechnology investment despite the existence of promising science. The 1983 Orphan Drug Act was passed to address this problem, offering financial incentives to companies developing rare disease drugs. During the first 25 years since the passage of the Act, 1,892 drugs were given orphan designation, 326 of which have been approved . While this astounding success has significantly impacted rare disease-affected patients, 95% of over 6,000 rare diseases still have no specific treatment.
Although there are numerous challenges in development of a disease treatment, the most critical is the clinical study process. By law, all drugs must undergo clinical trial testing to demonstrate safety and substantial efficacy before FDA approval. This process usually requires a Phase 3, double-blind, placebo-controlled trial, widely regarded as the "gold standard." Phase 3 trials typically assess efficacy using a measure of clinical benefit such as how a patient feels, functions, or survives . For disorders with very few and variable patients or lengthy time courses or irreversible disease progression, the use of clinical measures as endpoints can make specific treatment development intractable for practical and ethical reasons.
The Accelerated Approval (AA) regulations were promulgated by FDA in 1992 to drive the development of new treatments for serious and life-threatening disorders, primarily motivated by the AIDS crisis and the slow pace of treatment development for HIV infection . In HIV, monitoring death rate or complex endpoints such as hospitalization for opportunistic infections was difficult or unethical. Consequently, AA regulations allowed for drug approval based on the use of surrogate endpoints "reasonably likely to predict clinical benefit"  - a surrogate endpoint being a measure, such as a blood test or urine marker, believed to be indicative of a disease state and treatment effect, but not demonstrative of a direct health gain to the patient.
AA has been enormously successful at driving innovation in the development of cancer and HIV therapies during the first 16 years: 26 new chemical entity (NCE) cancer drugs have been approved, using either tumor load or progression-free survival as surrogate endpoints, and 29 HIV drugs (25 NCEs and 4 combination drugs) have gained approval using either CD4 count or viral load as endpoints . In the case of HIV, using survival as an endpoint would have made drug trials too time-consuming due to the lengthy disease course, too costly due to the high number of required patients, and unethical due to the use of placebo in a lethal condition. Even more critical for the development of HIV treatment was the ability to test the complex combinations of drugs responsible for the substantial improvement in long-term outcome. Drug combination studies would have been impossible using a clinical endpoint given large patient numbers and extended study lengths, despite the fact that drug combinations are now essential to the HIV treatment process. Current HIV therapies are a prime example whereby great scientific ideas would nonetheless have led to little or no treatment success without AA access. These successes show that the AA regulatory pathway is having a profound impact on innovation in treatment development.
Rare and ultra-rare diseases have not shared in cancer and HIV's successes. Only one drug among the 73 NCEs approved under AA, agalsidase beta (Fabrazyme ®), has been used to specifically treat a rare genetic disease, Fabry . Although treatments for biochemical genetic diseases, such as urea cycle disorders and phenylketonuria (PKU), have been approved through traditional FDA pathways using blood test endpoint-driven trials, the existence of regulatory precedents for approval (urea cycle drugs), substantial routine disease management history, and published study data (PKU drugs), supported the use of blood test-based endpoints [5, 6]. Most ultra-rare diseases do not have the same pre-existing body of clinical management or historical study data currently required to utilize the AA pathway. For example, during the clinical development of Aldurazyme to treat mucopolysaccharidosis I (MPS I), there was insufficient independent clinical data to support the use of the surrogate endpoint of urine glycosaminoglycan excretion to predict clinical benefit despite substantial scientific, animal, and clinical data supporting its relevance acquired in clinical trials. A second study using clinical endpoints was required for approval [7, 8]. With this essential requirement for independent clinical data, the AA pathway is virtually unavailable for novel drugs developed for untreated ultra-rare diseases.
The European Medicines Agency grants Conditional Marketing Authorization (CMA), similar to AA, to disease treatments addressing unmet medical need, including orphan diseases. Under CMA guidelines, treatments are conditionally approved on the basis of a presumed positive benefit-risk profile, and further confirmatory studies are agreed upon to establish clinical benefit. While the use of surrogate endpoints to obtain a CMA is not explicitly mentioned in legislation, industry has demonstrated a willingness to cooperate with government to make this a reality, and the necessity of collaboration between academia, government, and industry has been foreseen [9, 10]. Therefore, the use of surrogate endpoints to achieve drug approval is a pressing issue in more than one continent.
Three factors influencing the likelihood of initiation of a development program for rare disease treatment have been evaluated recently: prevalence, disease class, and scientific output. It is less than one-third as likely that a drug treating an ultra-rare disease with a very low prevalence of 0.1 to 9 per 100,000 enters the development process, as compared to a rare disease having a prevalence of 10 to 50 per 100,000 . The authors interpreted their finding as an indication that additional economic incentives and initiatives are necessary to promote ultra-rare disease treatment development.
We believe improving access to the AA pathway could potentially fulfill this need. By studying the effect of increased access to AA for ultra-rare diseases of a specified disease class and having ample scientific output, our analysis isolates one of the three factors studied by the authors: prevalence. We believe that all rare diseases, even those with more limited research activity or more poorly understood, would also benefit from this increased access to the AA pathway, given that a clearer path to approval will then drive more research in more productive directions.
To identify scientifically promising disease treatments that have not translated to human approved use and which might benefit from access to the AA pathway, we searched for ultra-rare disease therapies that consistently reversed disease pathology in animal models, but had stalled in development. We uncovered numerous examples of successful treatments, and investigated the probable effect that access to AA would have on investment potential for their development. After reviewing the diseases and their underlying science, we established reasonable clinical and surrogate endpoints and devised development programs dependent on both endpoint types. We demonstrate that improved AA access could enhance ultra-rare disease treatment feasibility and investment potential by decreasing the development cost to approval, shortening the time to approval, and increasing potential investment return. We are not verifying or implying that these surrogates have been proven to be valid, nor at this time are we suggesting methods to qualify surrogates more efficiently. By highlighting and quantifying the substantial benefit to development that we observe, we hope to spur experts in this field to discuss practical and rational solutions to the challenges of qualifying surrogates as primary endpoints for pivotal clinical trials in ultra-rare diseases.