Seeking Ways to Ensure Participant Safety in First-in-Human Trials

Kirsten Messmer, Ph.D., R.A.C., principal regulatory affairs specialist, details changes in regulatory guidelines for first-in-human trials.

In January 2016, the news of the hospitalization of five critically ill and one brain-dead patient dominated the media. The patients were clinical study volunteers in a first-in-human (FIH) trial in France. Ten years prior, six healthy volunteers in a study in London experienced severe adverse events during another FIH trial. Although severe adverse events or even deaths are extremely rare, both cases were potentially avoidable to some degree. Extensive investigations in both led to changes in regulatory guidelines, with the most recent being the update of European Medicines Agency (EMA) guidelines for FIH trials.

Let’s take a quick look at what happened in each of these two cases.

The 2016 incident occurred during the third arm (a multiple ascending dose arm) of a Phase I double-blind randomized placebo-controlled combined single and multiple ascending dose study including food interaction to investigate the safety, tolerability, pharmacokinetic and pharmacodynamics of BIA 10-2474. The compound was developed by Bial-Portela & Ca, and the trial was conducted by Biotrial, a French CRO. The volunteers experiencing adverse events received the second-highest dose under investigation and started experiencing symptoms of headaches, sight impairment and dizziness after the fifth dose. The first patient was hospitalized, fell into a coma and died a week later. Five additional patients received a sixth dose of the investigational product and were hospitalized shortly after the first with varying degrees of symptoms.

The 2006 incident occurred during the investigation of TGN-1412, a superagonistic anti-CD28 monoclonal antibody developed by TeGenero intended to balance the immune system in diseases such as rheumatoid arthritis and leukemia. The first group of eight volunteers received a single infusion — six receiving antibody and two placebo — with 10 minutes between each volunteer’s administration. Within 50 to 90 minutes all volunteers receiving the study drug experienced a severe cytokine storm syndrome resulting in multiple organ failure that required hospitalization. All affected volunteers survived, but will have long-term effects including disabilities and lost fingers and toes.

What did a review show?

Exhaustive investigations and in-depth reviews of data, trial procedures and events for both incidents concluded that the studies followed the protocols approved by the respective regulatory agencies and appropriate ethics committees, and that animal data did not predict the reactions observed in humans. A small molecular difference in a transmembrane domain between the cynomolgus macaques and humans failed to trigger the cytokine-storm observed with TGN-1412 in humans. Similarly, in the BIA 10-2474 study, the investigation following the incident showed that cerebral damage in mice, rats and primates was more extensive than initially thought. Most recent research using activity-based protein profiling confirmed an off-target effect of this particular investigational product disrupting lipid metabolism in the brain, which may have led to the brain damage. The researchers suggest the off-target effect might be species-specific and that more extensive testing in human cell and tissue models may have been able to indicate the possible toxicity.

At the time of the TeGenero incident, no clinical trial guidance addressing scientific and regulatory considerations for FIH trials existed apart from the Clinical Trial Directive 2001/20/EC (replaced by Regulation EU No 536/2014), the International Conference on Harmonisation (ICH) guidance on Good Clinical Practice (ICH E6) and ICH M3 addressing nonclinical data requirements and FIH dose determination. In 2006, the EMA released a concept paper regarding a planned guideline to support Phase I studies advocating approaches similar to those of the FDA’s guidance on exploratory IND studies supporting an expedited start of human trials. However, following the TeGenero incident, the EMA revised considerations in a guideline released in March 2007 recognizing that higher risk warranted better safeguards. Since the knowledge of investigational products at this early stage is still limited, remaining uncertainties create a risk to be addressed by study design. The guidance therefore suggested: Determining FIH dose by using the “minimum anticipated biological effect level” rather than the “no observed adverse event level” approach; requiring justification of non-sequential dosing; and including defined stopping rules. Identification and mitigation of risk were addressed in a further guidance later in 2007.

New guidance ahead?

In 2016, following the incidents during the Bial-Portela trial, the EMA again saw a need to update available guidance. Clinical trials have evolved in the past decade and protocols have become increasingly complex often involving multiple arms and combining several steps of clinical development into a single protocol. The 2007 guidance focused on the then typical single-ascending arm studies, while the current revision takes into account the increasing complexity of protocols and inclusion of multiple parts.

The revised guideline, released in July 2017, contains strategies for dose calculation, dose escalation and maximum dose in view of the more integral study approach. The guidance also focuses on improving the integration of nonclinical data, particularly pharmacokinetic, pharmacodynamics and toxicology data, into the assessment and mitigation. The importance of demonstrating the relevance of the animal model and assessing potential off-target effects is discussed in detail. Although nonclinical data supported the conduct of human studies in both cases, a more conservative approach should be considered in the future, as suggested by the more extensive toxicity seen during the in-depth investigations. The revised guidance also addresses in detail some of the operational aspects such as stopping rules, dose progression and rolling review of data, particularly in respect to safety information as it emerges during trial conduct. Both incidents involved operational practices that may have posed greater risk than good medical practice may have supported.

By their nature, Phase I clinical studies, and particularly FIH studies, pose a greater risk to volunteers due to the uncertainties that cannot be completely removed by nonclinical studies. The revised guidance aims to enhance strategies for risk identification and minimization to better ensure participant safety.

Kirsten Messmer, Ph.D., R.A.C., is a principal regulatory affairs specialist with the PPD™ clinical research business of Thermo Fisher Scientific.