Skip links

Single-Arm Trials Using Real-World Evidence for Rare Disease Product Development

As a sponsor designs clinical studies, the respective comparison control groups become a critical factor to consider. Often, to gain clinical trial design insights, a sponsor reviews the physician package inserts from approved New Drug Applications (NDAs) and Biologics License Applications (BLAs) with similar indications or in the same therapeutic area. If a single-arm trial was conducted as the pivotal trial for regulatory approval of those existing products, the sponsor will certainly want to do the same.

Similarly, in rare disease drug development, where a placebo control group may not be an option, a single-arm trial can be a successful approach. Premier Consulting has a breadth and depth of experience with single-arm trials, and this blog post shares some of the key insights we’ve learned.

What is the FDA’s current thinking on the use of single-arm trials?

While the FDA considers single-arm trials for regulatory approval, there are conditions that sponsors must meet. To approve a product, the FDA requires substantial evidence of effectiveness, consisting of adequate and well-controlled (A&WC) trials. As defined by the Code of Federal Regulations (CFR) §314.126, the purpose of conducting clinical investigations of a drug, biologic, or advanced therapy (gene or cell-based therapy) is to distinguish an active pharmaceutical ingredient’s effect from other influences, such as spontaneous change in the course of the disease, placebo effect, or biased observation.

Further, the CFR states that the clinical trial needs to use a design that permits a valid comparison with a control to provide a quantitative assessment of drug effect. This typically requires a concurrent placebo, an active control, a dose ranging control, or no treatment control. However, when a single-arm study is employed, the FDA recognizes a control group that incorporates real-world evidence (RWE) as a valid comparison—in other words, a natural history or historical control group. Natural history data becomes critical for product development when there is no concurrent control.

For more insights on RWE, see the FDA’s framework for RWE program.

What conditions make a single-arm study an A&WC trial to support regulatory applications?

At the 2020 World Orphan Drug Congress, Dr. Wilson Bryan, director of the FDA’s Office of Tissues and Advanced Therapies in the Center for Biologics Evaluation and Research (CBER), spoke about the features of a single-arm study that provides rigorous evidence of effectiveness. Premier Consulting’s experience suggests that the same conditions apply whether in CBER or in the Center for Drug Evaluation and Research (CDER):

  • The primary efficacy endpoint should be based on an outcome that does not occur (or seldom occurs) in nature.
  • The efficacy outcome should not be influenced by medical management and alternative therapies.
  • The primary endpoint assessment should be resistant to bias.
  • The effect size should be large when comparing the intervention group to the control group.

Oncology therapeutic products are good examples of where these A&WC conditions can be met. For example, we can know when a cancer goes into remission or is treated completely by a product, as this does not happen on its own and does not occur in nature. Single-arm studies are common in oncology, chiefly conducted in those patient populations that are relapsed and refractory after receiving other interventions. Single-arm trials are feasible in this case because the efficacy outcome is not influenced by medical management or alternative therapies.

A number of oncology products and advanced therapies have been approved with single-arm trials. Indirectly, the Hilal et al 2019 study examined 143 anticancer FDA drug approvals from 2013 to 2018, a number from which they excluded 47 approvals that were based on a single-arm study. In other words, a significant percentage of anticancer products approved in those five years gained approval using single-arm studies.

In addition, the primary endpoint assessment scale or tool should be resistant to bias introduction from either investigators or patients. Regulatory agencies want to know that the assessment measure is independent and cannot be influenced by the investigator or the patient’s own actions when evaluating the clinical efficacy outcome. The FDA has released a guidance on clinical trial endpoints to support effectiveness claims in NDAs and BLAs.

When a large effect size is observed between the intervention group and the control group, regulatory agencies believe that bias alone could not be responsible. If the primary endpoint assessment effect size observed is small when comparing the single-arm intervention group to the natural history group, regulators have a hard time assessing if substantial evidence has been demonstrated.

What is the top regulatory concern when a single-arm study is employed?

Simply, BIAS. As noted above, regulatory agencies are concerned with the introduction of bias within a single-arm study and its influence on results. There are several common sources of bias to consider:

  • Treatment assignments (intervention vs natural history group)
  • The patient
  • The investigator
  • Alternative treatments
  • The primary assessment tools or scales

Regulatory agencies want to ensure that a patient assigned to the intervention group is not treated differently from patients in the natural historical control group. One way to control for this is to limit the alternative treatments administered to the single-arm patients. It is easiest when no alternative therapies are available to impact the clinical disease course.

Regulatory agencies want to make sure that the clinical data collected from single-arm studies provides substantial evidence that the interventional product has the effect it purports to have, under the conditions of use recommended in labeling for patients with the disease state. The study design and primary analysis should minimize biases to the greatest extent possible to ensure the captured effect size is attributable to the interventional product and not to a confounder. Contact Premier Consulting today to find out how we can help you design an A&WC single-arm trial while minimizing bias introduction.


Ruth Stevens, PhD, MBA
Strategic Advisor