Considerations for Clinical Trial Design in Public Health Emergencies

While the fundamental concerns behind the design of most modern clinical trials are effectively the same, clinical trials that take place during public health emergencies face additional layers of challenges.

Timeliness of Clinical Trial Design in Emergencies        

The first and perhaps most challenging aspect of designing a clinical trial during an emergency is the issue of timeliness. Standard processes that provide for regulation and oversight of clinical trials, such as review by an investigational review board (IRB) and the FDA’s investigational new drug (IND) application process, tend to be lengthy by necessity. However, such processes are required even in the event of an emergency. During a potential pandemic, researchers and authorities must act on timelines which are orders of magnitude shorter than those seen in standard clinical trials. There are several existing mechanisms to address this conflict without compromising a study’s integrity or participant safety.

An investigator seeking to conduct research with a novel drug in a public health emergency must still operate under an IND. However, filing a new IND is a time and labor-intensive process which takes at least a month post-submission for approval. Ideally, an investigator or organization with a promising drug that can only be proven during an emergency will file an IND in advance. These INDs can be kept open without active ongoing studies in preparation for an emergency through close communication with the FDA and the continued submission of annual reports and other required regulatory materials. Original documents filed with the IND such as the protocol may be updated as necessary once more information is available.

However, this solution is not always feasible, as indications for investigational agents are not always conceived of before an emergency occurs. In these situations, investigators may choose to contact a drug’s manufacturer and determine if the drug can be made available under the company’s existing IND. If this is not possible, the FDA can issue an Emergency IND, authorizing the use of an experimental drug in an emergency without a standard full IND application.

The IRB review process is more difficult to streamline, as FDA regulations do not provide for terms like “expedited” or “compassionate” that imply that IRB approval can be accelerated. This problem is compounded by the fact that each hospital or research site has its own IRB, meaning that a multi-site study would need to wait on the review and approval of many individual IRBs. Both participant enrollment and treatment phases can be delayed during this process. To address this, FDA regulations allow for local IRBs to rely on all or some of the review findings of a reputable centralized IRB (CIRB). Under this review model, CIRBs and IRBs may agree upon cooperative review arrangements to avoid lengthy, duplicative, and resource-intensive reviews. Approval from the CIRB in a multi-site study would apply to all sites simultaneously. Of course, IRBs and CIRBs must work together to clarify the ultimate responsibilities of each party, and to avoid scenarios where an investigator is compliant with FDA regulations but noncompliant with their own institution’s regulations.

In situations where the primary concern is to provide clinical care, the FDA may approve an Emergency Use Authorization (EUA), which would allow for the use of a product without IRB approval. It is important to note that data collected from patients treated under an EUA is not acceptable for product approval. Examples of active FDA-approved EUAs in 2015 include assays and tests for Ebola, H7N9, Middle East Respiratory Syndrome Coronavirus (MERS-CoV), and oral formulations of doxycycline products to treat inhalational anthrax.

Investigators, sponsors, and institutional authorities can also employ their own strategies to help speed up the regulatory process, independent of the FDA. These include using open repositories of protocols or protocol parts that have been pre-screened for ethical issues, convening review teams in person, using remote meeting technology to facilitate international review of protocol materials, or conducting rolling reviews of protocol portions as they are completed. Preparing pre-emergency, pre-screened FDA submissions that can be tailored to developing outbreaks can allow the FDA to review and assess materials prior to emergencies, allowing for the quick turnaround.

Once IND and IRB requirements are met, practical steps can be taken to determine whether additional resources are required to complete site initiations, staff trainings, transportation risk assessments, supply chain design, or other complex tasks on a tight timeline. In 2001, TRI’s efforts under the Centers for Disease Control and Prevention (CDC) to develop the anthrax absorbed vaccine during the 2001 anthrax bioterrorism threat included distributing study essential documents, developing regulatory documents associated with the IND, and initiating a protocol in less than 24 hours.

Ethics of Clinical Trial Design in Emergencies

The urgency in public health emergencies inextricably links many timeliness issues with ethical ones. Even the standard informed consent process becomes very tricky if the patient requires immediate care but is unresponsive and unaccompanied by their legal representative. FDA regulations provide for scenarios such as these; in this case, if the criteria of 21 CFR 50.23(a) are met and certified by the investigator and a physician who is not participating in the clinical investigation, the requirement of informed consent can be waived.

In more urgent cases, an investigator may consider immediately administering a test article if the same criteria are met but another physician is unavailable, as long as retroactive follow-up requirements (such as notification of the IRB and retroactive review by another physician) are met. During large scale, life-threatening emergent situations, the regulations in 21 CFR 50.24 provide for study-wide waiving of informed consent. In such scenarios, a research plan must be approved in advance by FDA and the IRB, and publically disclosed to the community in which the research will be conducted.

The race against time does not affect only the investigator or study designers, but also the prospective patient who is under great pressure to secure treatment. Traditionally, a patient outside of a study may take experimental drugs through three routes: their physician’s therapeutic privilege, through compassionate “named patient” laws, and through use of parallel track agents. Consenting to take part in a conventional study may prove to be an inferior option when weighed against these other options. From the perspective of a participant, enrolling in a study that is recruiting patients who are vulnerable or infected with a fast-acting and/or lethal disease and being assigned to a placebo arm may have a massive opportunity cost.

Because of this concern, investigators designing experiments in the face of an epidemic must decide if the golden standard for clinical research, the placebo-controlled trial, is appropriate. In recent years, an increasingly popular alternative is the use of active controls, for example in the form of wedged cluster designs. In this study design, participants would be grouped into clusters, usually by geographical location, and receive treatment in staggered intervals. Those waiting for the treatment act as the controls. The benefits of this type of trial design are that all participants receive treatment, and investigators still have controls to compare to their study agent data.

But as is often the case, one solution to an ethical problem raises other problems. Using a cluster design means that some clusters of patients will receive the study agent before others. Participants who must wait to receive a study agent may lose valuable treatment time as their diseases progress, or as they continue to face exposure. In developing countries, this issue may take on a socioeconomic flavor, since investigators are more likely to begin treatment with clusters based in locations with existing infrastructure. There is a real risk that staggered clinical trials that appear to prioritize one group of people over another may ignite latent social inequality issues in the region. This compounds the fact that low-resource areas tend to have worse treatment options in the case of an epidemic in the first place, either due to a lack of appropriate facilities or a low assigned priority in the allocation of resources.

Unintentionally contributing to social unrest in a host country undermines one of the goals of foreign assistance, which is to build goodwill in other countries. In especially poor regions, researchers must take care to not take advantage of economically or educationally disadvantaged persons who cannot give true informed consent. To address these issues, researchers must include host countries in the clinical design process.

Validity of Clinical Trial Design in Emergencies

Even after timing and ethical considerations are taken into account, clinical trials in public health emergencies must be diligent in their record-keeping and data management practices. Emergency measures implemented during a crisis will be subject to intense scrutiny afterwards. For example, during the SARS outbreak in 2003, ribavirin and corticosteroids were administered to thousands of patients outside of a formal clinical study, leaving physicians and investigators without efficacy data to consult for their own recommendations in the event of a future outbreak. This missing data could have also been used to track participant welfare post-intervention. The H1N1 vaccine Pandemrix, administered to patients in northern Europe during the 2009 swine flu epidemic, has recently been linked to an increased risk of developing narcolepsy. Most recently, a nasal spray vaccine for swine flu was found to be unusually sensitive to heat, losing potency when being prepared for administration. This knowledge is essential for improving treatment options in the event of a recurrence, for example in the current H1N1 outbreaks in India and Saudi Arabia.

This demonstrates the importance of keeping statistical considerations in mind throughout the course of the clinical trial. Study agent efficacy and safety data collected during a rare outbreak will be needed in the event of subsequent incidents, but is often spotty and incomplete. Accrual may be difficult to predict or control during a public health emergency, while retention for long-term evaluation may be difficult if the disease is lethal. Good data collection and management practices can quickly take a back seat to the urgency of treating patients.

Public health emergencies add a great deal of difficulty to the administration and execution of clinical trials. Experience and flexibility are vital in navigating the various challenges that may arise.

Jeffry Liu is Clinical Operations Specialist for Technical Resources International, Inc. tech-res.com.