Navigating the complexity of CGT clinical trial design

As increasing numbers of cell and gene therapies (CGTs) — a method of treating disease through the use of genes, tissues or cells — are approved by regulatory bodies and made available to patients, so too are the numbers of CGT treatments in sponsor pipelines on the rise. According to the Alliance for Regenerative Medicine, there are currently more than 1,200 companies developing CGT therapies. This technology offers the hope of new treatments for cancer and many rare diseases, and is expanding into applications in indications that impact a greater portion of the population.

However, CGT faces several obstacles in clinical trials that more traditional therapies may not. This blog will explore some of the challenges of CGT clinical trial design, such as patient recruitment and data collection, and strategies for how to approach them.

Patient recruitment and retention

Patient recruitment and retention is a common challenge faced by CGT trials. Many rare disease trials are, by definition, forced to recruit from an extremely limited pool of patients, who tend to be geographically diverse and are often paediatric patients with life-threatening conditions. Meanwhile, in indications that affect a broader subset of the population, other treatments may already exist, and there may be more competition for trial participation. With an approach that is still largely hypothetical, and that may hold risks that are yet unknown, CGT trial recruitment must fight an uphill battle. Moreover, sponsors must bear in mind the need to consent to long-term follow-up (LTFU) of up to 15 years’ duration (as per US FDA Guidance) in CGT trials, as and when required.

One option for sponsors is to partner with patient advocacy groups. Contacting a group centred around a trial’s condition of interest can help to locate patients, as well as professionals and sites with related expertise. This can be particularly useful in the case of rare diseases, as patients are not always easily identified via electronic health records (EHRs). Additionally, engaging with advocacy groups can help to build awareness about a particular study.

New analytics capabilities also offer possibilities for recruitment. Tools have been developed that can automatically sift through EHRs to determine potential study participants. As technology progresses, a variety of these tools are becoming available, such as those that make use of machine learning to find and analyse text surrounding key words in electronic records,¹ or are embedded in the EHR to notify physicians that a patient meets trial criteria and to facilitate referral at the point of care.²

An approach that has become more normalised due to the limitations imposed by the COVID-19 pandemic is decentralised or hybrid trials. These trials minimise the need for participants to physically visit a central trial site and therefore reduce many of the barriers that make it difficult to recruit and retain patients. For example, travel to a site can be difficult for patients dealing with severe illness. And in the case of rare diseases, where participants may be very spread out, it can be hard to ensure there is a site nearby for all patients resulting in participants needing travel often long distances. Options such as virtual visits and electronic apps that allow patients to report their outcomes can facilitate remote monitoring and data collection in the patients’ own homes, and wearable devices can gather data such as a patient’s heart rate, respiration or activity. However, in many cases safety concerns for CGT treatment means that it is important to have ready access to medical assistance. Where appropriate, the use of home health care may still help to ease some of the patients’ burden.

Data collection

As established above, the sample size for CGT trials may be limited, thereby also limiting the amount of data it is possible to collect. This means it is imperative to make the best possible use of time and data. Adaptive trials, which establish a plan for data review and alterations to the trial as it is underway, can help to make better use of resources and, in some cases, require fewer participants. Some examples of adaptive designs include: sample size re-estimation, which adjusts sample size to ensure the desired power; population enrichment, which focuses recruitment of those who will most benefit from the treatment; and continual reassessment method, which employs model-based dose escalation to determine the maximum tolerated dose.³

Related to the model-based continual reassessment method are Bayesian trial designs. Using a trial design that utilises Bayesian statistics allows the integration of past knowledge, such as preclinical data or data gained from previous cohorts, with any new data gained. This can help sponsors get as much statistically sound information as possible from the data generated in a clinical trial.

Looking forward

There has been rapid growth in the area of Cell and Gene Therapy development over the last few years which holds incredible potential for life changing treatments for people suffering from serious diseases. In order to drive a future in which these therapies are seen more commonly within standard of care, we need to start laying the groundwork now to navigate some of the obstacles along the road ahead.

References:

1. https://ascopubs.org/doi/full/10.1200/CCI.20.00180
2. https://www.fiercehealthcare.com/health-tech/new-ehr-tool-automates-patient-clinical-trial-referrals-providers
3. https://bmcmedicine.biomedcentral.com/articles/10.1186/s12916-018-1017-7/tables/1