Author: Kazem Kazempour, Ph.D.

As the world slowly begins to emerge from the COVID-19 pandemic, some patients continue to struggle with symptoms of the virus months after their initial infection. From fatigue and dyspnea to chest discomfort, cough and loss of taste and/or smell, these patients have been unable to return to full health. One recent trial found that roughly 25% of COVID-19 patients experience prolonged illness beyond the typical three-week period.1 Another found that 35% of patients who had a positive reverse transcription-polymerase chain reaction (RT-PCR) test reported not having returned to their normal state of health after 14-21 days.²

Those who have continued to experience symptoms for months — a condition referred to here as “post-acute COVID-19” and known by various other names, including “long-haul COVID,” “long COVID syndrome” and “chronic COVID-19” — remain a primary concern for health care providers and pharmaceutical companies alike. Although therapies are in development and trials for potential treatments are underway, many unknown variables remain. The wide array of symptoms makes it difficult to identify patients, there is only limited data regarding the relationship between symptom severity and duration, and post-acute COVID-19 itself lacks a clear definition. From the standpoint of trial design and endpoints, this scenario parallels the AIDS/HIV pandemic in the 1980s and early 1990s, when the virus’s behavior was not well known and narrowing the research focus — with the only real clinical endpoint being the prolongation of life — became its own separate challenge.

As before, this environment complicates clinical trials, from initial design through to analysis, potentially adding time and costs to the process when time is already of the essence. Despite challenges, however, it is possible to design an effective trial by applying some of what was learned during the AIDS/HIV epidemic and considering each phase of the trial concurrently. Taking this approach, rather than working more linearly with a fixed randomized controlled trial (RCT), allows you to anticipate and respond to data, become agile, and more easily accommodate new information while remaining grounded in science and not losing the integrity of the clinical trial.

Designing for Efficiency While Including Every Possible Variable

Previous trial data and clinical endpoints are already helping researchers shape new trials for post-acute COVID-19 syndrome. By looking at previous data addressing different conditions with similar symptoms, including chronic fatigue syndrome (CFS), and familiar endpoints, such as the six-minute walk test (6MWT), researchers are developing their studies with an existing foundation.³ However, as was learned in the early AIDS/HIV trials, clinical endpoints may not always be the best option.

Initially, prolongation of life appeared to be the only reasonable endpoint with AIDS/HIV research, as deciding between factors such as viral load, weight loss and other potential options was challenging. Quickly, many involved in the process, from research to approval, realized the patient mortality endpoint required too much time, significantly delaying research efforts as patients remained desperate for help. To overcome this challenge, researchers identified a surrogate endpoint, CD4 cell counts, which would provide a clinically objective measurement more efficiently. This level of adaptability was inarguably one of the main factors that helped encourage significant progress in AIDS/HIV research, and it paved the way for the use of surrogate markers in other clinical trials. Given the significant variability in symptoms and patient experience with post-acute COVID-19, a clinical endpoint may not be readily available. However, if a surrogate endpoint can be identified across symptoms — such as brain fog or anything that clinicians and regulators can agree on — researchers could potentially see results more quickly.

Similarly, although adaptive clinical trials (ACTs) were not part of the clinical landscape during AIDS/HIV research, this approach is also proving beneficial today. To maintain accuracy and compliance with guidelines from the U.S. Food and Drug Administration (FDA) or other regulatory bodies outside of the United States, clinical trials cannot change as they progress, but they can be designed to account for potential modifications in advance of conducting the trial. Adaptive clinical trials, also known as adaptive trial designs, allow researchers to build anticipated changes into the design in order to improve accuracy, respond to key learnings in the moment with science and without losing the integrity of the trial, work with an optimized trial population, strengthen trial ethics, gather stronger results and ultimately increase the chances of approval from sponsors.⁴ In other words, ACTs allow you to better work with and through a post-acute COVID-19 trial without adding months to your timeline. This is clearly different from the process of identifying a surrogate marker, but ACTs also demonstrate the importance of applying science to find new ways of building more effective trials.

An ACT creates the ability to make modifications, provided that these modifications are predefined, follow a predetermined set of guidelines and are based on a carefully executed, ongoing analysis of the trial data. This means researchers can use previous trial data and endpoints as guides to better understand the various paths their trial may take. ACTs can also be confirmatory or exploratory in design, allowing you to investigate hypotheses while remaining responsive to trial data.⁵ In all cases, the trial is designed with both conduct and analysis in mind to ensure a cohesive, continuous approach.

Scientific Accuracy Remains the Pivot Point

Executing predetermined changes requires a shift in how you conduct the research and understand interim analyses, meaning that the agility of the research team extends from the design process through to the trial itself. These changes must also be implemented while maintaining trial integrity and without derailing the process. Any number of things can occur during a trial, and the research team needs to remain flexible to adjust according to predefined specifications and gather the most accurate and valuable data in pursuit of approval. This responsiveness is essential and depends entirely on the experience and flexibility of the team.

In addition to the variability of post-acute COVID-19 symptoms making the syndrome itself difficult to define, trial conduct remains challenging due largely to patient recruitment. Without knowing the biological or immunological mechanisms for the syndrome or what makes some people more susceptible than others, proper screening is complex, and it is difficult to identify the best target for a potential therapy.⁶ Lingering symptoms tend to affect older patients, those with comorbidities such as diabetes and respiratory disease, and those who have had more serious cases of COVID-19, but this is not exclusively the case. Post-acute COVID-19 is also seen in younger patients and after more mild cases, and these patients may not present if they do not make the connection between their symptoms and the syndrome.

Moreover, early data suggests that COVID-19 vaccinations may provide symptom relief for some long-haul patients, while approximately 10-15% report feeling worse after vaccination.⁷ This adds additional complications because the reasons for these responses are still unknown and vaccination could change a patient’s profile after screening. However, this landscape also further emphasizes the importance of an agile team and an adaptive design. As the trial progresses, an ACT allows you to shift patient populations to focus on a specific age or symptom, efficiently direct recruitment efforts as early as possible, and respond appropriately as data comes to light.

Turning Ongoing Analysis Into Comprehensive — and Compliant — Findings

Having everything running in parallel from the very beginning leaves you better prepared at the conclusion of the trial if every step was completed with precision and experience. Rather than working sequentially, from design to analysis, look and work ahead. Adaptive design makes this possible by offering clarify from the start. Full recruitment needs, including potential standby recruitment, can be identified, reducing timelines to a matter of days rather than weeks. Analysis methodologies can be developed before the data are collected, allowing researchers to quickly view the results and provide them to the advisory committee. Unnecessary bureaucracy stands in the way of rapid trial execution, eliminating flexibility and leading to delays, but potential challenges present themselves regularly.

Delayed data between study sites, new or revised regulatory procedures, or something as unforeseen as a global pandemic that separates teams and keeps trial subjects at home can quickly dissolve years of work and planning if coordination is lacking. Similarly, these challenges can delay efforts to engage in new trials, including those related to post-acute COVID-19. As we have learned through the pandemic, remaining endlessly responsive and precisely coordinated is the best way to overcome and adapt to the unexpected.

This level of flexibility, and the understanding required to know when to adapt a trial, how to execute that adaptation and how it will affect the data, is the result of extensive experience not only with research, but with ACTs. Expert biostatisticians begin planning around logistical challenges while designing the trial, justify every response with data, and conduct complicated interpretations to make both in-trial decisions and post-trial analysis as simple as possible. These efforts combine to keep post-acute COVID-19 trials on target and help the research deliver value.

Engaging a research partner or an experienced contract research organization (CRO) can help you embrace fluidity, from design and conduct to analysis and approval, especially when working through an ACT. As the health care community continues to learn more about post-acute COVID-19, patient populations become more well defined, and the FDA responds in kind with revised guidelines and expectations, responsiveness and scientific accuracy remain essential to building a greater understanding and uncovering an effective treatment. Working with an experienced team alleviates strain on your own team and frees up internal resources to focus on continued drug development and other priorities. Most importantly, this partnership will keep your trials on track, bringing your development efforts closer to providing patients with relief from their lingering symptoms.

References:

1. Greenhalgh, T. (2020, August 11). Management of post-acute covid-19 in primary care. The BMJ. www.bmj.com/content/370/bmj.m3026

2. Tenforde, M. W. (2020, July 30). Symptom Duration and Risk Factors for Delayed Return to Usual HealthAmong Outpatients with COVID-19 in a Multistate Health Care Systems Network — United States, March-June 2020. Centers for Disease Control and Prevention. www.cdc.gov/mmwr/volumes/69/wr/mm6930e1.htm

3. Vaidya, M., & Newton, W. (2021, February 11). Trials of Covid-19 long-hauler therapies could repurpose endpoints from other disease studies but evolving definitions pose challenges, experts say. Clinical Trials Arena. www.clinicaltrialsarena.com/comment/trials-of-covid-19-long-hauler-therapies-could-repurpose-endpoints-from-other-disease-studies

4. NSF International. (2020, October 12). White Paper: Understanding Adaptive Designs for Clinical Trials. www.nsf.org/knowledge-library/white-paper-understanding-adaptive-designs-for-clinical-trials

5. Bhatt, D. L., & Mehta, C. (2016). Adaptive Designs for Clinical Trials. The New England Journal of Medicine, 375(1), 65-74. doi.org/10.1056/nejmra1510061

6. Studies elucidate poorly understood long COVID. (2021, July 16). CIDRAP. www.cidrap.umn.edu/news-perspective/2021/07/studies-elucidate-poorly-understood-long-covid