Rare diseases are defined in the Orphan Drug Act as diseases or conditions that affect less than 200,000 people in the United States. There are over 7,000 rare diseases affecting more than 30 million people in the United States, and despite the FDA’s approval of over 600 treatments for rare diseases since signing the Orphan Drug Act into law in 1983, most rare diseases still do not have a treatment. Evaluating and developing treatments for rare diseases remains a key priority for FDA and is incentivized through the Orphan Drug Act. Sponsors granted orphan drug designation qualify for the following incentives:
Despite financial incentives, the regulatory burden of Sponsors developing orphan products is the same as with all drugs and biologics in that substantial evidence of efficacy and safety in the specific indication must be demonstrated.
Development of drugs and biologics intended to treat orphan diseases is often accomplished in small clinical trial(s). Clinical trials for rare diseases are often complicated by a number of factors, including small, heterogeneous patient populations, poor understanding of the natural history of the disease, and identifying an appropriate control group.
Standard clinical study designs often fail to demonstrate statistical significance in small and heterogenous patient populations. Specifically, with small sample sizes, P values are particularly susceptible to slight variations in the observed number of outcomes. Commentary by Mitani and Haneuse (2020) provides an example of this in their summary of 125 patient study on hyperthyroidism, where the estimated prevalence in the study population was 1 of 125 patients (0.8%) and not found to be statistically different than the general population prevalence (0.5%), with a P value of 0.88. However, the P value would have been considerably different if 2 of 125 patients (P =0.23) or 3 of 125 patients (P = 0.04) would have been diagnosed with hyperthyroidism in the study. This example shows that if just two additional cases of hyperthyroidism had been observed in the study, the standard P value threshold associated with statistical significance (i.e., P < 0.05) would have been achieved. Likewise, genotypic and phenotypic heterogeneity in many rare diseases may introduce nondrug-related variability that hinders the ability of the clinical study to demonstrate a potential treatment effect. Additionally, disease heterogeneity may create a situation in which no single efficacy endpoint is suitable to measure clinical benefit in the trial.
In the case of many rare diseases, information on the natural history of the disease is limited, which leads to inefficient study design and/or improper selection of endpoints. Information on the natural history of the disease provides an understanding of the typical trajectory of the disease in the absence of intervention or with standard of care treatment. Understanding the natural history of the disease is critical to designing a well-controlled clinical trial with meaningful and relevant endpoints. Specifically, natural history studies can emphasize what clinical benefit may look like in the patient population or provide support for developing an appropriate surrogate biomarker. Although FDA does not require Sponsors to conduct natural history studies, FDA’s thinking on the utility of natural history studies is captured in the March 2019 Draft Guidance for Industry – Rare Diseases: Natural History Studies for Development where FDA describes the use of natural history studies in identifying the patient population, identifying or developing clinical outcome assessments, identifying or developing biomarkers, and designing externally controlled studies.
In the development of drugs and biologics, it is well understood that randomized, double-blind, placebo-controlled trials are the gold standard in meeting FDA’s criteria of demonstrating substantial evidence of effectiveness through adequate and well-controlled investigations. However, in the development of products for rare diseases, it may not be feasible to conduct a randomized, placebo-controlled study, which can complicate the interpretation of results regarding the effect of the investigational product. As such, FDA has detailed two strategies to minimize unknown variables that could affect the outcome independent of the intervention. These strategies include using historical (external) controls and early randomization in clinical development. Additionally, alternative study designs including adaptive designs and crossover trial designs may be appropriate to investigate the safety and efficacy of treatments for rare diseases.
Determining an appropriate patient population, meaningful endpoints, and acceptable randomization and control strategies for clinical trials of rare diseases requires collaboration between clinicians, statisticians, patients/advocacy groups, investigators, and regulatory agencies. Through formal meetings with the FDA, Sponsors can obtain feedback on the proposed clinical study design and seek the Agency’s agreement that the proposed investigation will likely produce the necessary data to support a marketing application.
Mitani AA and Haneuse S. 2020. Small data challenges of studying rare diseases. JAMA Netw Open. 3(3):e201965.